GeekZoneBooks.Com: Guide to Wireless Networking

© 2007 GeekZoneBooks.Com

GeekZoneBooks.Com

Guide to Wireless Networking

Travel through the history of computer networking and learn how computers and cell phones send data over the air in a secure manner from the moon and back.

Forward

Convergence … In the very near future all media will be converged on mobile wireless hand held devices.

On-Demand Video & Audio, Live Streaming Video & Audio, Voice, Picture Messaging, E-Mail and Internet Access, Games, GPS Mapping Applications, Digital Key Lock, Bar Code Processing, Bio-Metric Id. Security, Personal Information Managers, payment processing settlement services and other yet to be developed applications.

The FCC has set a goal for broadband coverage of speeds over 400Kbs on all wireless devices that will allow the 3G and 4G revolution to happen. (Updated Info: It took 12Mbps to make this happen)

So if you’re looking at this on a wireless device you are holding in your hands the kind of device that has connected you to the future of convergence.

At the end of convergence all the world’s knowledge will be able to be accessed from your data phone communicator.

The satellites will carry the information to all parts of the world and the headset will be able to communicate with any other device on the network. Towers will no longer be needed to carry radio wave based communications.

Each device will able to carry the person’s entire history and important documentation and only be accessible by that person because of biometric identification methods.

They will be able to do real time language conversion on all voice communications and textual information.

Brew — (the language of cell phone applications) or its future incarnations will morph into a ubiquitous open standard that the platform of applications will be built on. (Updated Info: Android won the open source platform and iOS runs on Apple IPhones)

Data compression will mature so full motion video and audio receive and transmit will be possible.

The user of the devices will always be at a disadvantage always because of the amount of research they would have to do to understand the foundations of the operation of the industry so Informatics companies such as WDS Global will always have a niche value to the networks to provide support for end users.

When will convergence be complete? When the global satellite communication system is in place and tested and accepted by the populace as economically affordable for service and device acquisition.

The system must support the bandwidth to provide the room for IPV6 to have every possible IP number available assigned to a device and allow full motion transmit/receive in high-def digital audio/video formats.

When this system is designed and in place then the device industry will be able to design the hand held units that will allow the end user to interface with the network.

Perhaps the interim phase will be a hybrid cell tower/satellite/fiber system with the satellites carrying the back haul bandwidth and using the cell towers and fiber to the door to distribute the communications to handset units.

This is what I see and what I prognosticate. Where will you be and what part will you play in this inevitability? Be a part of history, don’t just read about it.

Michael Scott McGinn

March 21, 2007

Introduction: The history of networking

Because networking is such a broad and complex subject, no single event represents its point of origin. However, we can arbitrarily decide on the 1960’s as the formative period for the field, since that is when the computer began to affect significantly the lives of ordinary individuals and the day-to-day operation of both business and government.

1960’s

In the 1960s computer networking was essentially synonymous with mainframe computing, and the distinction between local and wide area networks did not yet exist. Mainframes were typically “networked” to a series of dumb terminals with serial connections running on RS—232 or some other electrical interface. If a terminal in one city needed to connect with a mainframe in another city, a 300- baud long—haul modem would use the existing analog Public Switched Telephone Network (PSTN) to form the connection. The technology was primitive indeed, but it was an exciting time nevertheless.

To continue the story, the quality and reliability of the PSTN increased significantly in 1962 with the introduction of pulse code modulation (PCM), which converted analog voice signals into digital sequences of bits. A consequent development was the first commercial touch-tone phone, which was introduced in 1962. Before long, digital phone technology became the norm, and DS-O (Digital Signal Zero) was chosen as the basic 64-kilobit—per-second (Kbps) channel upon which the entire hierarchy of the digital telephone system was built. A later development was a device called a channel bank, which took 24 separate DS-O channels and combined them together using time-division multiplexing (TDM) into a single 1.544—Mbps channel called DS-1 or T1. (In Europe, 30 DS-O channels were combined to make E1.) When the backbone of the Bell telephone system finally became fully digital years later, the transmission characteristics improved significantly for both voice and data transmission due to higher quality and less noise associated with Integrated Services Digital Network (ISDN) digital lines, though local loops have remained analog in many places. But that is getting a little ahead of the story.

The first communication satellite, Telstar, was launched in 1962. This technology did not immediately affect the networking world because of the high latency of satellite links compared to undersea cable communications, but it eventually surpassed transoceanic underwater telephone cables (which were first deployed in 1965 and could carry 130 simultaneous conversations) in carrying capacity. In fact, early in 1960 scientists at Bell Laboratories transmitted a communication signal coast-to-coast across the United States by bouncing it off the moon! By 1965 popular commercial communication satellites such as Early Bird were being widely deployed and used.

The year 1969 witnessed an event whose full significance was not realized until more than two decades later: namely, the development of the ARPANET packet-switching network. ARPANET was a project of the US. Department of Defense’s Advanced Research Projects Agency (ARPA), which became DARPA in 1972. Similar efforts were underway in France and the United Kingdom, but it was the US project that eventually evolved into the present-day Internet. (France’s MlNTEL packet-switching system, which was based on the X25 protocol and which aimed to bring data networking into every home, did take off in 1984 when the French government started giving away MlNTEL terminals; by the early 1990s, more than 20 percent of the country’s population was using it.) The original ARPANET network connected computers at Stanford University, the University of California at Los Angeles (UCLA), the University of California at Santa Barbara (UCSB), and the University of Utah, with the first node being installed at UCLA’s Network Measurement Center. A year later, Harvard University, the Massachusetts Institute of Technology (MIT), and a few other prominent institutions were added to the network, but few of those involved could imagine that this technical experiment would someday profoundly affect modern society and the way we do business.

That same year, Bell Laboratories developed the UNIX operating system, a multitasking, multi-user network operating system (NOS) that became popular in academic computing environments in the 1970s. A typical UNIX system in 1974 was a PDP—11 minicomputer with dumb terminals attached. In a configuration with 768 kilobytes (KB) of magnetic core memory and a couple of ZOO-megabyte (MB) hard disks, the cost of such a system would have been around $40,000. Many important standards for computer systems also evolved during the 1960s. In 1962, IBM introduced the first 8-bit character encoding system, called Extended Binary-Coded Decimal Interchange Code (EBCDIC). A year later the competing American Standard Code for Information Interchange (ASCII) was introduced. ASCII ultimately won out over EBCDIC even though EBCDIC was 8-bit and ASCII was only 7—bit. The American National Standards Institute (ANSI) formally standardized ASCII in 1968. ASCII was first used in serial transmission between mainframe hosts and dumb terminals in mainframe computing environments, but it was eventually extended to all areas of computer and networking technologies.

Other developments in the 1960s included the development in 1964 of IBM’s powerful System/360 mainframe computing environment, which was widely implemented in government, university, and corporate computing centers. In 1966, IBM introduced the first disk storage system, which employed 50 metal platters, each of which was 2 feet (0.6 meter) wide and had a storage capacity of 5 MB. IBM created the first floppy disk in 1967. In 1969, Intel Corporation released a RAM chip that stored 1 KB of information, which at the time was an amazing feat of engineering.

1970’s

Although the 1960’s were the decade of the mainframe, the 19703 gave birth to Ethernet, which today is by far the most popular LAN technology. Ethernet was born in 1973 in Xerox Corporation’s research lab in Palo Alto, California. (An earlier experimental network called ALOHAnet was developed in 1970 at the University of Hawaii.) The original Xerox networking system was known as X-wire and worked at 2.94 Mbps. X—wire was experimental and was not used commercially, although a number of Xerox Palo Alto workstations used for word processing were networked together in the White House using X-wire during the Carter administration. In 1979, Digital Equipment Corporation (DEC), Intel, and Xerox formed the DIX consortium and developed the specification for standard 10-Mbps Ethernet, or thicknet, which was published in 1980. This standard was revised and additional features were added in the following decade.

The conversion of the backbone of the Bell telephone system to digital circuitry continued during the 1970s and included the deployment in 1974 of the first digital data service (DDS) circuits (then called the Dataphone Digital Service). DDS formed the basis of the later deployment of ISDN and T1 lines to customer premises, and AT&T installed its first digital switch in 1976.

In wide area networking, a new telecommunications service called X.25 was deployed toward the end of the decade. This new system was packet-switched, in contrast to the circuit-switched PSTN, and later evolved into public X25 networks such as GTE’s Telenet Public Packet Distribution Network (PDN), which later became SprintNet. X.25 was widely deployed in Europe, where it still maintains a large installed base, especially for communications in the banking and financial industry.

In 1970 the Federal Communications Commission (FCC) announced the regulation of the fledgling cable television industry. Cable TV remained primarily a broadcast technology for delivering entertainment to residential homes until the mid-1990s, when technologies began to be developed to enable it to carry broadband services to residential subscribers. Cable modems now compete strongly with Digital Subscriber Line (DSL) as the main two forms of broadband Internet access technologies.

Despite all these technological advances, however, telecommunications services in the 1970s remained unintegrated, with voice, data, and entertainment carried on different media. Voice was carried by telephone, which was still analog at the customer premises; entertainment was broadcast using radio and television technologies; and data was usually carried over RS-232 or Binary Synchronous Communication (BSC) serial connections between dumb terminals and mainframes (or, for remote terminals, long-haul modem connections over analog telephone lines).

The 1970s were also notable for the growth of ARPANET, which grew throughout the decade as additional hosts were added at various universities and government institutions. By 1971 the network had 19 nodes, mostly consisting of a mix of PDP-8, PDP-11, IBM 8/360, DEC-10, Honeywell, and other mainframe and minicomputer systems linked together. The initial design of ARPANET called for a maximum of 265 nodes, which seemed like a distant target in the early 1970s. The initial protocol used on this network was NCP, but this was replaced in 1982 by the more powerful TCP/IP protocol suite. In 1975 the administration of ARPANET came under the authority of the Defense Communications Agency.

ARPANET protocols and technologies continued to evolve using the informal RFC process developed in 1969. In 1972 the Telnet protocol was defined in RFC 318, followed by FTP in 1973 (RFC 454). ARPANET became an international network in 1973 when nodes were added at the University College of London in the United Kingdom and at the Royal Radar Establishment in Norway. ARPANET even established an experimental wireless packet-switching radio service in 1977, which two years later became the Packet Radio Network (PRN ET).

Meanwhile, in 1974 the first specification for the Transmission Control Protocol (TCP) was published. Progress on the TCP/IP protocols continued through several iterations until the basic TCP/lP architecture was formalized in 1978, but it was not until 1983 that ARPANET started using TCP/IP instead of NCP as its primary networking protocol.

The year 1977 also saw the development of UNIX to UNIX Copy (UUCP), a protocol and tool for sending messages and transferring files on UNIX-based networks. An early version of the USENET news system using UUCP was developed in 1979. (The Network News Transfer Protocol [NNTP] came much later, in 1987.)

In 1979 the first commercial cellular phone system began operation in Japan. This system was analog in nature, used the 800-MHz and 900-MHz frequency bands, and was based on a concept developed in 1947 at Bell Laboratories.

An important standard to emerge in the 1970s was the public—key cryptography scheme developed in 1976 by Whitfield Diffie and Martin Hellman. This scheme underlies the Secure Sockets Layer (SSL) protocol developed by Netscape Communications, which is still the predominant approach for ensuring privacy and integrity of financial and other transactions over the World Wide Web (WWW). Without SSL, popular e-business sites such as Amazon and eBay would have a hard time attracting customers!

Among other miscellaneous developments during this decade, in 1970 IBM researchers invented the relational database, a set of conceptual technologies that has become the foundation of today’s distributed application environments. In 1971, IBM demonstrated the first speech recognition technologies–which have since led to those annoying automated call handling systems found in customer service centers! IBM also developed the concept of the virtual machine in 1972 and created the first sealed disk drive (the Winchester) in 1973. In 1974, IBM introduced the Systems Networking Architecture (SNA) for networking its mainframe computing environment. In 1971, Intel released its first microprocessor, a 4-bit processor called the 4004 that ran at a clock speed of 108 kilohertz (kHz), a snail’s pace by modern standards but a major development at the time. Another significant event was the launching of the online service CompuServe in 1979, which led to the development of the first online communities.

The first personal computer, the Altair, went on the market as a kit in 1975. The Altair was based on the Intel 8080, an 8-bit processor, and came with 256 bytes of memory, toggle switches, and light-emitting diode (LED) lights. Although the Altair was basically for hobbyists, the Apple II from Apple Computer, which was introduced in 1977, was much more. A typical Apple II system, which was based on the Motorola 6502 8-bit processor, had 4 KB of RAM, a keyboard, a motherboard with expansion slots, built-in BASIC in ROM, and color graphics. The Apple II quickly became the standard desktop system in schools and other educational institutions. However, it was not until the introduction of the IBM Personal Computer (PC) in 1981 that the full potential of personal computers began to be realized, especially in businesses.

In 1975, Bill Gates and Paul Allen licensed their BASIC computer programming language to MITS, the Altair’s manufacturer. BASIC was the first computer language specifically written for a personal computer. Gates and Allen coined the name “Microsoft” for their business partnership, and they officially registered it as a trademark the following year. Microsoft Corporation went on to license BASIC to other personal computing platforms such as the Commodore PET and the TRS-80.

1980’s

In the 1980s the growth of client/server LAN architectures continued while that of mainframe computing environments declined. The advent of the IBM PC in 1981 and the standardization and cloning of this architecture led to an explosion of PC-based LANs in businesses and corporations around the world, particularly with the release of the IBM PC AT hardware platform in 1984. The number of PCs in use grew from 2 million in 1981 to 65 million in 1991. Novell, which appeared on the scene in 1983, soon became a major player in file and print servers for LANs with its Novell NetWare platform.

However, the biggest development in the area of LAN networking in the 1980s was the continued evolution and standardization of Ethernet. While the DIX consortium worked on Ethernet standards in the late 1970s, the lEEE with its Project 802 initiative tried working toward a single unified LAN standard. When it became clear that this goal was impossible, Project 802 was divided into a number of separate working groups, with 802.3 focusing on Ethernet, 802.4 on Token Bus, and 802.5 on Token Ring technologies and standards. The work of the 802.3 group resulted in the first Ethernet standard, called 10Base5 or thicknet, which was almost identical to the version developed by DIX. 10Base5 was called thicknet because it used thick coaxial cable, and in 1985 the 802.3 standard was extended to include 10Base2 using thin coaxial cable, commonly called thinnet.

Through most of the 1980s, coaxial cable was the main form of cabling used for implementing Ethernet. A company called SynOptics Communications, however, developed a product called LattisNet that was designed for transmitting 10-Mbps Ethernet over twisted—pair wiring using a star-wired topology that was connected to a central hub or repeater. This wiring was cheaper than coaxial cable and was similar to the wiring used in residential and business telephone wiring systems. LattisNet was such a commercial success that in 1990 the 802.3 committee approved a new standard called 10BaseT for Ethernet that ran over twisted-pair wiring. 10BaseT soon superseded the coaxial forms of Ethernet because of its ease of installation and because its hierarchical star-wired topology was a good match for the architectural topology of multistory buildings.

In other Ethernet developments, fiber-optic cabling, first developed in the early 1970s by Corning, found its first commercial networking application in Ethernet networking in 1984. (The technology itself was standardized as 1OBaseFL in the early 1990s.) In 1988 the first fiber—optic transatlantic undersea cable was laid and greatly increased the capacity of transatlantic communication systems.

Ethernet bridges became available in 1984 from DEC and were used both to connect separate Ethernet LANs to make large networks and to reduce traffic bottlenecks on overloaded networks by splitting them into separate segments. Routers could be used for similar purposes, but bridges generally offered better price and performance, as well as less complexity, during the 1980s. Again, market developments preceded standards, as the IEEE 802.1D Bridge Standard, which was initiated in 1987, was not standardized until 1990.

In the UNIX arena, the development of the Network File System (NFS) by Sun Microsystems in 1985 resulted in a proliferation of diskless UNIX workstations having built—in Ethernet interfaces. This development helped drive the demand for Ethernet and accelerated the evolution of Ethernet bridging technologies into today’s switched networks. By 1985 the rapidly increasing numbers of UNIX hosts and LANs connected to the ARPANET began to transform it from what had been mainly a network of mainframe and minicomputer systems into something

like what it is today. The first UNIX implementation of TCP/IP came in v4.2 of Berkeley’s BSD UNIX, from which other vendors such as Sun Microsystems quickly ported their versions of TCP/IP. Although PC-based LANs rapidly grew in popularity in business and corporate settings during the 1980s, UNIX continued to dominate in academic and professional high-end computing environments as the mainframe environment declined.

IBM introduced its Token Ring networking technology in 1985 as an alternative LAN technology to Ethernet. IBM had submitted its technology to the IEEE in 1982 and the 802.5 committee standardized it in 1984. IBM soon supported the integration of Token Ring with its existing SNA networking services and protocols for IBM mainframe computing environments. The initial Token Ring specifications delivered data at 1 Mbps and 4 Mbps, but it dropped the 1-Mbps version in 1989 when it introduced a newer 16-Mbps version. Interestingly, no formal IEEE specification exists for 16—Mbps Token Ring–vendors simply adopted IBM’s technology for the product. Efforts were made to develop high-speed Token Ring, but these have finally been abandoned and today Ethernet reigns supreme.

Also in the field of local area networking, in 1982 the American National Standards Institute (ANSI) began standardizing the specifications for Fiber Distributed Data Interface (FDDI). FDDI was designed to be a high-speed (100 Mbps) fiber-optic networking technology for LAN backbones on campuses and industrial parks. The final FDDI specification was completed in 1988, and deployment in campus LAN backbones grew during the late 1980s and the early

1990s. But today FDDI is considered legacy technology and has been superseded in most places by Fast Ethernet and Gigabit Ethernet (GbE).

In 1983 the ISO developed an abstract seven-layer model for networking called the Open Systems Interconnection (OSI) reference model. Although some commercial networking products were developed based on OSI protocols, the standard never really took off, primarily because of the predominance of TCP/IP. Other standards from the ISO and ITU that emerged in the 1980s included the X.400 electronic messaging standards and the X500 directory recommendations, both of which held sway for a while but have now largely been superseded X.4OO by the Internet’s Simple Mail Transfer Protocol (SMTP) and X.5OO by Lightweight Directory Access Protocol (LDAP).

A major event in the telecommunications/WAN field in 1984 was the divestiture of AT&T as the result of the seven-year antitrust suit brought against AT&T by the US. Justice Department. AT&T’s 22 Bell operating companies were formed into 7 new RBOCs (only 4 are left today). This meant the end of the old Bell telephone system, but these RBOCs soon formed the Bellcore telecommunications research establishment to replace the defunct Bell

Laboratories. The United States was then divided into Local Access and Transport Areas (LATAs), with intra-LATA communication handled by local exchange carriers (the Bell Operating Companies or B003) and inter-LATA communication handled by interexchange carriers (IXCs) such as AT&T, MCI, and Sprint Corporation.

The result of the breakup was increased competition, which led to new WAN technologies and generally lower costs. One of the first effects was the offering of T1 services to subscribers in 1984. Until then, this technology had been used only for backbone circuits for long-distance communication. New hardware devices were offered to take advantage of the increased bandwidth, especially high-speed T1 multiplexers, or muxes, that could combine voice and data in a single communication stream. The year 1984 also saw the development of digital Private Branch Exchange (PBX) systems by AT&T, bringing new levels of power and flexibility to corporate subscribers.

The Signaling System #7 (SS7) digital signaling system was deployed within the PSTN in the 1980s, first in Sweden and later in the United States. 887 made new telephony services available to subscribers, such as caller ID, call blocking, and automatic callback.

The first trials of ISBN, a fully digital telephony technology that runs on existing copper local loop lines, began in Japan in 1983 and in the United States in 1987. All major metropolitan areas in the United States have since been upgraded to make ISDN available to those who want it, but ISDN has not caught on in the United States as a WAN technology as much as it has in Europe.

The 1980s also saw the standardization of SONET technology, a high-speed physical layer (PHY) fiber-optic networking technology developed from time— division multiplexing (TDM) digital telephone system technologies. Before the divestiture of AT&T in 1984, local telephone companies had to interface their own TDM-based digital telephone systems with proprietary TDM schemes of long-distance carriers, and incompatibilities created many problems. This provided the impetus for creating the SONET standard, which was finalized in 1989 through a

series of Comité Consultatif International Télégraphique et Téléphonique (CCITT; anglicized as International Telegraph and Telephone Consultative Committee) standards known as G.707, G.608, and G.709. By the mid-1990s almost all long-distance telephone traffic in the United States used SONET on trunk lines as the physical interface.

The 1980s brought the first test implementations of Asynchronous Transfer Mode (ATM) high-speed cell-switching technologies, which could use SONET as the physical interface. Many concepts basic to ATM were developed in the early 1980s at the France-Telecom laboratory in Lannion, France, particularly the PRELUDE project, which demonstrated the feasibility of end-to-end ATM networks running at 62 Mbps. The CCiTT standardized the 53—byte ATM cell format in 1988, and the new technology was given a further push with the

creation of the ATM Forum in 1991. Since then, use of ATM has grown significantly in telecommunications provider networks and has become a high-speed backbone technology in many enterprise-level networks around the world. However, the vision of ATM on users’ desktops has not been realized because of the emergence of cheaper Fast Ethernet and GbE LAN technologies and because of the complexity of ATM itself.

The convergence of voice, data, and broadcast information remained a distant vision throughout the 1980s and was even set back because of the proliferation of networking technologies, the competition between cable and broadcast television, and the slow adoption of residential ISDN. New services did appear, however, especially in the area of commercial online services such as America Online (AOL), CompuServe, and Prodigy, which offered consumers e-mail, bulletin board systems (8883), and other services.

A significant milestone in the development of the Internet occurred in 1982 when the networking protocol of ARPANET was switched from NCP to TCP/IP. On January 1, 1983, NCP was turned off permanently——anyone who had not migrated to TCP/lP was out of luck. ARPANET, which connected several hundred systems, was split into two parts, ARPANET and MILNET.

The first international use of TCP/lP took place in 1984 at the Conseil Européen pour la Recherche Nucléaire (CERN), a physics research center located in Geneva, Switzerland. TCP/lP was designed to provide a way of networking different computing architectures in heterogeneous networking environments. Such a protocol was badly needed because of the proliferation of vendor—specific networking architectures in the preceding decade, including “homegrown” solutions developed at many government and educational institutions. TCP/IP

made it possible to connect diverse architectures such as UNIX workstations, VMS minicomputers, and Cray supercomputers into a single operational network. TCP/lP soon superseded proprietary protocols such as Xerox Network Systems (XNS), ChaosNet, and DECnet. It has since become the de facto standard for internetworking all types of computing systems.

CERN was primarily a research center for high-energy particle physics, but it became an early European pioneer of TCP/IP and by 1990 was the largest subnetwork of the Internet in Europe. In 1989 a CERN researcher named Tim Berners-Lee developed the Hypertext Transfer Protocol (HTTP) that formed the basis of the World Wide Web (WWW). And all of this developed as a sidebar to the real research that was being done at CERN–slamming together protons and electrons at high speeds to see what fragments would appear!

Also important to the development of Internet technologies and protocols was the introduction of the Domain Name System (DNS) in 1984. At that time, ARPANET had more than 1000 nodes and trying to remember their numerical lP addresses was a headache. DNS greatly simplified that process. Two other internet protocols were introduced soon afterwards: NNTP was developed in 1987, and Internet Relay Chat (lRC) was developed in 1988.

Other systems paralleling ARPANET were developed in the early 1980s, including the research—oriented Computer Science NETwork (CSNET), and the Because It’s Time NETwork (BITNET), which connected IBM mainframe computers throughout the educational community and provide email services. Gateways were set up in 1983 to connect CSNET to ARPANET, and BITNET was similarly connected to ARPANET. In 1989, BITNET and CSNET merged into the Corporation for Research and Educational Networking (CREN).

In 1986 the National Science Foundation NETwork (NSFNET) was created. NSFNET networked the five national supercomputing centers together using dedicated 56-Kbps lines. The connection was soon seen as inadequate and was upgraded to 1.544-Mbps T1 lines in 1988. In 1987, NSF and Merit Networks agreed to jointly manage the NSFNET, which had effectively become the backbone of the emerging Internet. By 1989 the Internet had grown to more than 100,000 hosts, and the Internet Engineering Task Force (lETF) was officially created to administer its development. In 1990, NSFNET officially replaced the aging ARPANET and the modern Internet was born, with more than 20 countries connected.

Cisco Systems was one of the first companies in the 1980s to develop and market routers for Internet Protocol (IP) internetworks, a business that today is worth billions of dollars and is a foundation of the Internet. Hewlett-Packard was Cisco’s first customer for its routers, which were originally called gateways.

In wireless telecommunications, analog cellular was implemented in Norway and Sweden in 1981. Systems were soon rolled out in France, Germany, and the United Kingdom. The first US. commercial cellular phone system, which was named the Advanced Mobile Phone Service (AMPS) and operated in the 800- MHz frequency band, was introduced in 1983. By 1987 the United States had more than 1 million AMPS cellular subscribers, and higher-capacity digital

cellular phone technologies were being developed. The Telecommunications Industry Association (TIA) soon developed specifications and standards for digital cellular communication technologies.

A landmark event that was largely responsible for the phenomenal growth in the PC industry (and hence the growth of the client/server model and local area networking) was the release of the first version of Microsoft’s text-based, 16-bit MS-DOS operating system in 1981. Microsoft, which had become a privately held corporation with Bill Gates as president and chairman of the board and Paul Allen as executive vice president, licensed MS—DOS version 1 to IBM for its PC. MS-DOS continued to evolve and grow in power and usability until its final version, MS-DOS 6.22, which was released in 1993. One year after the first version of MS- DOS was released in 1981, Microsoft had its own fully functional corporate network, the Microsoft Local Area Network (MILAN), which linked a DEC 206, two PDP-11/7OS, a VAX 11/250, and a number of MC68000 machines running XENIX. This type of setup was typical of the heterogeneous computer networks that characterized the early 1980s.

In 1983, Microsoft unveiled its strategy to develop a new operating system called Windows with a graphical user interface (GUI). Version 1 of Windows, which shipped in 1985, used a system of tiled windows that allowed users to work with several applications simultaneously by switching between them. Version 2 was released in 1987 and supported overlapping windows and support for expanded memory.

Microsoft launched Its SQL Server relational database server software for LANs in 1988. In its current version, SQL Server 2000 is an enterprise-class application that competes with other major database platforms such as Oracle and DB2. IBM and Microsoft jointly released their 32-bit OS/2 operating system in 1987 and released 08/2 1.1 with Presentation Manager a year later.

In miscellaneous developments, IBM researchers developed the Reduced Instruction Set Computing (RISC) processor architecture in 1980. Apple Computer introduced its Macintosh computing platform in 1984 (the successor of its Lisa system), which introduced a windows-based GUI that was the precursor to Windows. Apple also introduced the 3.5-inch floppy disk in 1984. Sony Corporation and Philips developed CD-ROM technology in 1985. (Recordable CD-R technologies were developed in 1991.) IBM released its AS/400 midrange computing system in 1988, which continues to be popular to this day.

1990’s

The 1990s were an explosive decade in every aspect of networking, and we can only touch on a few highlights here. Ethernet continued to evolve as a LAN technology and began to eclipse competing technologies such as Token Ring and FDDI. In 1991, Kalpana Corporation began marketing a new form of bridge called a LAN switch, which dedicated the entire bandwidth of a LAN to a single port instead of sharing it among several ports. Later known as Ethernet switches or Layer 2 switches, these devices quickly found a niche in providing dedicated

high-throughput links for connecting servers to network backbones. Layer 3 switches soon followed, eventually displacing traditional routers in most areas of enterprise networking except for WAN access. Layer 4 and higher switches are now popular in server farms for load balancing and fault tolerance purposes.

The rapid evolution of the PC computing platform and the rise of bandwidth- hungry applications created a need for something faster than 10-Mbps Ethernet, especially on network backbones. The first full-duplex Ethernet products, offering speeds of 20 Mbps, became available in 1992. In 1995 work began on a standard for full—duplex Ethernet; it was finalized in 1997. A more important development was Grand Junction Networks’ commercial Ethernet bus,

introduced in 1992, which functioned at 100 Mbps. Spurred by this advance, the 802.3 group produced the 802.3u 100BaseT Fast Ethernet standard for transmission of data at 100 Mbps over both twisted-pair copper wiring and fiber—optic cabling.

Although the jump from 10-Mbps to 100-Mbps Ethernet took almost 15 years, a year after the 100BaseT Fast Ethernet standard was released, work began on a 1000-Mbps version of Ethernet popularly known as Gigabit Ethernet (GbE). Fast Ethernet was beginning to be deployed at the desktop, and this was putting enormous strain on the FDDI backbones that were deployed on many commercial and university campuses. FDDI also operated at 100 Mbps (or 200 Mbps if fault tolerance was discarded in favor of carrying traffic on the redundant ring), so a single Fast Ethernet desktop connection could theoretically saturate the capacity of the entire network backbone. Asynchronous Transfer Mode (ATM), a broadband cell-switching technology used primarily in telecommunication/WAN environments, was briefly considered as a possible successor to FDDI for backboning Ethernet networks together, and LAN emulation (LANE) was developed to carry LAN traffic such as Ethernet over ATM. However, ATM is much more complex than Ethernet, and a number of companies saw extending Ethernet speeds to 1000 Mbps as a way to provide network backbones with much greater capacity using technology that most network administrators were already familiar with. As a result, the 802 group called 802.32 developed a GbE standard called 1000BaseX, which it released in

1998. Today GbE is the norm for LAN backbones, and Fast Ethernet is becoming ubiquitous at the desktop level. Work is even unden/vay on extending Ethernet technologies to 10 gigabits per second (Gbps). A competitor of GbE for high- speed collapsed backbone interconnects, called Fibre Channel, was conceived by an ANSI committee in 1988 but is used mainly for storage area networks (SANs).

The 1990s saw huge changes in the landscape of telecommunications providers and their services. “Convergence” became a major buzzword, signifying the combining of voice, data, and broadcast information into a single medium for delivery to businesses and consumers through broadband technologies such as metropolitan Ethernet, Digital Subscriber Line (DSL), and cable modern systems. The cable modem was introduced in 1996, and by the end of the decade broadband residential Internet access through cable television systems had become a strong competitor with telephone-based systems such as Asymmetric Digital Subscriber Line (ADSL) and G.Lite, another variant of DSL.

Also in the 1990s, Voice over IP (VoiP) emerged as the latest “Holy Grail” of networking and communications and promised businesses huge savings by routing voice telephone traffic over existing IP networks. VoiP technology works, but the bugs are still being ironed out and deployments remain slow. Recent developments in VoiP standards, however, may help propel deployment of this technology in coming years.

The first public frame relay packet-switching services were offered in North America in 1992. Companies such as AT&T and Sprint installed a network of frame relay nodes across the United States in major cities, where corporate networks could connect to the service through their local telco. Frame relay began to eat significantly into the deployed base of more expensive dedicated leased lines such as the T1 or E1 lines that businesses used for their WAN solutions, resulting in lower prices for these leased lines and greater flexibility of services.

In Europe frame relay has been deployed much more slowly, primarily because of the widespread deployment of packet—switching networks such as X.25.

The Telecommunications Act of 1996 was designed to spur competition in all aspects of the US. telecommunications market by allowing the RBOCs access to long-distance services and lXCs access to the local loop. The result has been an explosion in technologies and services offered by new companies called competitive local exchange carriers (CLECs), with mergers and acquisitions changing the nature of the service provider landscape almost daily.

The 1990s saw a veritable explosion in the growth of the internet and the development of Internet technologies. As mentioned earlier, ARPANET was replaced in 1990 by NSFNET, which by then was commonly called the internet. At the beginning of the 1990s, the internet’s backbone consisted of 1.544-Mbps T1 lines connecting various institutions, but in 1991 the process of upgrading these lines to 44.735-Mbps T3 circuits began. By the time the Internet Society (ISOC) was chartered in 1992, the Internet had grown to an amazing 1 million hosts on almost 10,000 connected networks. In 1993 the NSF created Internet Network Information Center (interNIC) as a governing body for DNS. In 1995 the NSF stopped sponsoring the Internet backbone and NSFNET went back to being a research and educational network. Internet traffic in the United States was routed through a series of interconnected commercial network providers.

The first commercial Internet service providers (ISPs) emerged in the early 1990s when the NSF removed its restrictions against commercial traffic on the NSFNET. Among these early ISPs were Performance Systems International (PSI), UUNET, MCI, and Sprintlink. (The first public dial-up ISP was actually The World, with the URL www.world.std.com.) In the mid-1990s, commercial online networks such as AOL, CompuServe, and Prodigy provided gateways to the Internet to subscribers. Later in the decade, Internet deployment grew exponentially, with personal Internet accounts proliferating by the tens of millions around the world, new technologies and services developing, and new paradigms evolving for the economy and business. It would take a whole book to talk about all the ways the Internet has changed our lives.

Many Internet technologies and protocols have come and gone quickly. Archie, an FTP search engine developed in 1990, is hardly used today. The WAIS protocol for indexing, storing, and retrieving full-text documents, which was developed in 1991, has been eclipsed by Web search technologies. Gopher, which was created in 1991, grew to a worldwide collection of interconnected file systems, but most Gopher servers have now been turned off. Veronica, the Gopher search tool developed in 1992, is obviously obsolete as well. Jughead later supplemented Veronica but has also become obsolete. (There never was a Betty.)

The most obvious success story among Internet protocols has been HTTP, which, with HTML and the system of URLs for addressing, has formed the basis of the Web. Tim Berners-Lee and his colleagues created the first Web server (whose fully qualified DNS name was info.cern.ch) and Web browser software using the NeXT computing platform that was developed by Apple pioneer Steve Jobs. This software was ported to other platforms, and by the end of the decade more than 6 million registered Web servers were running, with the numbers growing rapidly.

Lynx, a text-based Web browser, was developed in 1992. Marc Andreessen developed mosaic, the first graphical Web browser, in 1993 for the UNIX X Windows platform while he was a student at the National Center for Supercomputing Applications (NCSA). At that time, there were only about 50 known Web servers, and HTTP traffic amounted to only about 0.1 percent of the

Internet’s traffic. Andreessen left school to start Netscape Communications, which released its first version of Netscape Navigator in 1994. Microsoft Internet Explorer 2 for Windows 95 was released in 1995 and rapidly became Netscape Navigator’s main competition. In 1995, Bill Gates announced Microsoft‘s wide-ranging commitment to support and enhance all aspects of Internet technologies through innovations in the Windows platform, including the popular Internet

Explorer Web browser and the Internet Information Server (IIS) Web server platform of Windows NT. Another initiative in this direction was Microsoft’s announcement in 1996 of its ActiveX technologies, a set of tools for active content such as animation and multimedia for the Internet and the PC.

In cellular communications technologies, the 1990s were clearly the “digital decade.” The work of the TIA resulted in 1991 in the first standard for digital cellular communication, the TDMA Interim Standard 54 (IS-54). Digital cellular was badly needed because the analog cellular subscriber market in the United States had grown to 10 million subscribers in 1992 and 25 million subscribers in 1995. The first tests of this technology, based on Time Division Multiple Access (TDMA) technology, took place in Dallas, Texas, and in Sweden, and were a success. This standard was revised in 1994 as TDMA IS-136, which is commonly referred to as Digital Advanced Mobile Phone Service (D-AMPS).

Meanwhile, two competing digital cellular standards also appeared. The first was the CDMA iS-95 standard for CDMA cellular systems based on spread spectrum technologies, which was first proposed by QUALCOMM in the late 1980s and was standardized by the TlA as IS-95 in 1993. Standards preceded implementation, however; it was not until 1996 that the first commercial CDMA cellular systems were rolled out.

The second system was the Global System for Mobile Communication (GSM) standard developed in Europe. (GSM originally stood for Groupe Spéciale Mobile.) GSM was first envisioned in the 1980s as part of the movement to unify the European economy, and the European Telecommunications Standards Institute (ETSI) determined the final air interface in 1987. Phase 1 of GSM deployment began in Europe in 1991. Since then, GSM has become the predominant system for cellular communication in over 60 countries in Europe, Asia, Australia, Africa, and South America, with over 135 mobile networks implemented. However, GSM implementation in the United States did not begin until 1995.

In the United States, the FCC began auctioning off portions of the 1900-MHz frequency band in 1994. Thus began the development of the higher-frequency Personal Communications System (PCS) cellular phone technologies, which were first commercially deployed in the United States in 1996.

Establishment of worldwide networking and communication standards continued apace in the 1990s. For example, in 1996 the Unicode character set, a character set that can represent any language of the world in 16-bit characters, was created, and it has since been adopted by all major operating system vendors.

In client/server networking, Novell in 1994 introduced Novell NetWare 4, which included the new Novell Directory Services (NDS), then called NetWare Directory Services. NDS offered a powerful tool for managing hierarchically organized systems of network file and print resources and for managing security elements such as users and groups. NetWare is now in version 6 and NDS is now called Novell eDirectory.

In other developments, the U.S. Air Force launched the twenty-fourth satellite of the Global Positioning System (GPS) constellation in 1994, making possible precise terrestrial positioning using handheld satellite communication systems. Real Networks released its first software in 1995, the same year that Sun Microsystems announced the Java programming language, which has grown in a few short years to rival C/C++ in popularity for developing distributed applications. Amazon.com was launched in 1995 and has become a colossus of cyberspace retailing in a few short years. Microsoft WebTV, introduced in 1997, is beginning to make inroads into the residential Internet market.

Finally, the 1990s were, in a very real sense, the decade of Windows. No other technology has had as vast an impact on ordinary computer users as Windows, which brought to homes and workplaces the power of PC computing and the opportunity for client/server computer networking. Version 3 of Windows, which was released in 1990, brought dramatic increases in performance and ease of use over earlier versions, and Windows 3.1, released in 1992, quickly became the standard desktop operating system for both corporate and home users. Windows for Workgroups 3.1 quickly followed that same year. It integrated networking and workgroup functionality directly into the Windows operating system, allowing Windows users to use the corporate computer network for sending email, scheduling meetings, sharing files and printers, and performing other collaborative tasks. In fact, it was Windows for Workgroups that brought the power of computer networks from the back room to users’ desktops, allowing them to perform tasks previously possible only for network administrators.

In 1992, Microsoft released the first beta version of its new 32-bit network operating system, Windows NT. In 1993 came MS—DOS 6, as Microsoft continued to support users of text-based computing environments. That was also the year that Windows NT and Windows for Workgroups 3.11 (the final version of 16-bit Windows) were released. In 1995 came the long-awaited release of Windows 95, a fully integrated 32-bit desktop operating system designed to replace MS-DOS, Windows 3.1, and Windows for Workgroups 3.11 as the mainstream desktop operating system for personal computing. Following in 1996 was Windows NT 4, which included enhanced networking services and a new Windows 95-style user interface. Windows 95 was superseded by Windows 98 and later by Windows Millennium Edition (Me).

2000’s

At the turn of the millennium came the long-anticipated successor to Windows NT, the Windows 2000 family of operating systems, which includes Windows 2000 Professional, Windows 2000 Server, Windows 2000 Advanced Server, and Windows 2000 Datacenter Server. The Windows family has how grown to encompass the full range of networking technologies, from embedded devices and Personal Digital Assistants (PDAs) to desktop and laptop computers to heavy-duty servers running the most advanced, powerful, scalable, business-critical, enterprise-class applications.

In 2000 Windows XP was released and this was a complete makeover for the masses. It had a more appealing user interface and came bundled with advanced accessories as standard tools. it had a video-editing program called Movie Maker and it came with a build in Web Server that could be activated and configured to allow each user to become a publisher of content for the Internet. From 2000 to 2007 XP was the dominant desktop operating system for the masses and reached to a 90% market penetration level. Microsoft Office also gained the dominant spot for end user office software, which included a Word Processor, Spreadsheet, Relational Database Management System, Presentation Creation Software, and a Web Site Editing Program.

This decade was also the rise of Linux on the desktop, which is still gaining market share. Using creative applications such as Open Office enhanced compatibility between Linux and Windows file structure. Linux is gaining on the market share strong hold that Microsoft enjoys. Linux was created by Linus Trovalds ( http://en.wikipedia.org/wiki/Linus_Torvalds )at the University of Helsinki to replace Unix. Linus wanted an open source operating system that could be created by a community of developers who would create programs that could be freely distributed under the GPL software license. The GPL allowed anyone to copy the program and add to it and then repackage and redistribute the resulting program as long as the GPL license was provided to the program and the source code was included to allow the next person to do the same. This GPL license was fundamentally different from the Closed Source Copyright that Microsoft and other commercial software developers used to protect their intellectual property and control the modifications and distribution of the software. This is the reason that Microsoft was able to grow so large and so fast.

Today the Open Source VS Closed Source battle continues to be fought in the trenches of the marketplace. Both systems have merits and benefits and ultimately achieve the same results which are providing usable operating systems to end users to allow them to be productive. Linux based web servers enjoy a larger share of the market then Windows based web servers because the Linux based operating system was more affordable and in many cases completely free so it was widely implemented by small and large web site operators to gain entry to the internet

2007 also saw the introduction of the virtual computer or virtual appliance. This development allowed a computer running Windows XP to run a virtual computer inside of the memory and load in a Linux virtual appliance and run that appliance in a window on the system without affecting the host operating system. This allowed casual users to test out Linux applications and see them run without installing a full Linux operating system and erasing the host operating system. We also saw the creation of multi-core processors and multi processor motherboards which greatly enhanced the processing power of a single pc box.

The future of networking is bright as faster processors and fatter faster data pipes are created as well as new compression and transmission protocols are approved and released into the marketplace. It is truly an interesting time to be in the high technology field of networking.

For more information on this subject and to keep up with new and exciting additions to this topic visit our web site at http://wwwgeekzonehosting.com

The History of Wireless Phones

The Mobile Lifecycle

The mobile phone is the preferred voice device for hundreds of millions of users. The main reason is that a handset is personal: in your possession 24 hours a day, 7 days a week. It contains important information conveniently; including contacts, and may also include diary information and notes. It can be used to access the Internet, especially for information and email. in this section I will look at the development of cellular radio technology since its first inception to the present day and beyond.

Way back when Mobile communication services delivered using cellular radio were first introduced over twenty years ago: development began in the late 1960s and the first mobile networks were launched in the early 1980s. Before the days of cellular networks people who desperately needed mobile communications could use a vehicle-mounted radio telephone which was serviced by one large central antenna in a major city. These antennas could provide 25 channels at best and the vehicle transmitter needed to be powerful enough to transmit over distances up to 50 miles.

First Generation Systems (1 G)

These first mobile networks were entirely analogue and are referred to as “first generation”, or 1G, systems. Initially the number of users was comparatively low and growth projections modest. Most countries where mobile service was available had just one operator that was state-owned. Co-ordination between countries was minimal and roaming between countries with the same handset was impossible. Several different technologies emerged:

The United States opted for the AMPS system (Advanced Mobile Phone System).

The UK implemented TACS (Total Access Communications System).

The Scandinavian countries implemented NMTS (Nordic Mobile Telephone System).

Due to the differences in spectral allocation between these countries (the way in which the radio spectrum was allocated for different uses), these cellular technologies worked on different frequencies.

(AMPS is still widely used in North and South America, especially in rural areas, and I will therefore look at it briefly later on.)

Devices were bulky with limited battery life, and connections and coverage area were, at best, adequate. Little provision was made for security and the first mobiles were subject to a practice known as “cloning”. Anyone with some spare time and some spare cash could isolate a phone’s individual identity code and program it into another handset, hence gaining free access to the network at some poor individual’s expense. These first analogue networks also had little to

no capacity to handle data communications. This is not to say that these networks could not transmit data, but initially there was little demand for such services.

Second Generation Systems (2G)

Second generation systems were launched in the early 1990s and were characterized by the implementation of digital technology. Perhaps the biggest catalyst for digitalization came in 1982, when the European Conference of Posts and Telecommunications Administrations (CEPT), consisting then of the telecommunication administrations of twenty six nations, created a working group to develop a set of standards for a pan—European cellular network — the Groupe Spéciale Mobile (the origin of the acronym GSM). The goal was to adopt and develop a single technology for the Single European Market. The new digital technology would replace the existing patchwork of incompatible analogue systems and offer European citizens the opportunity to make and receive calls anywhere in Europe. Digital technology also provided improved voice quality, security and data-capability at 9600 bits per second.

The first GSM (Global System for Mobile Telecommunications, as the acronym now stands for) networks were publicly launched in Europe in 1992. Meanwhile, the Americans also adopted digital technology. Due to the prevalence of AMPS, it was not cost efficient to implement a new GSM system, and so existing AMPS infrastructure was upgraded to D-AMPS (Digital Advanced Mobile Phone System).

The D-AMPS system is also referred to as the TDMA (Time Division Multiple Access) system, as this is the technology that is employed to increase network capacity. Read the later sections for an examination of the technologies involved in modern and future cellular radio communications.

Due to its efficient use of the spectrum of frequency allocated to use by cellular radio, GSM technology allowed more operators to enter the market and hence service more subscribers. A second band was licensed for use by GSM networks by the European regulating body to secure investment confidence, drive competition and provide more capacity.

GSM networks were rolled out in North America, but whilst Europe and Asia were adopting it unilaterally, initial deployment in the US was slow. Another digital technology was also available which offered much higher capacity capabilities than GSM and data rates at 14.4 Kbps — this technology was CDMA (Code Division Multiple Access). Japan’s NTT DoCoMo also shunned GSM and adopted an entirely proprietary technology called PDC (Packet Digital Communications) which is unique to Japan.

These, then, are the four digital cellular radio technologies within the 2G definition: GSM, TDMA, CDMA and PDC.

2.5G Systems

Despite the plethora of improvements these 2G digital systems heralded, the network operators soon realized that they would soon be outpaced by the growth in popularity of mobile telephone communications. It also became clear that the modern subscriber required a much broader range of services, much of them data-related, than they were in a position to provide. They therefore decided to implement a number of enhancements.

These enhancements are known as 2.5G systems as they are effectively “bolt- ons” to the existing network infrastructure: a stop-gap to accommodate the flood of new subscribers while network operators can roll out 3G systems.

GSM operators implemented GPRS (General Packet Radio Service) as well as, in the case of Orange, HSCSD (High Speed Circuit-Switched Data).

D-AMPS / TDMA operators in North America implemented CDPD (Cellular Digital Packet Data)

CDMA operators in North America improved the existing air interface to provide data rates of up to 19.2 Kbps

This brings us to the present day…

Today

178 countries worldwide now embrace GSM, serving a truly mass market of 677 million users (as at the end of March 2002). GSM is the only cellular system that is standardized for use in all four cellular bands worldwide (850, 900, 1800 and 1900 MHz). The GSM standard has evolved over a decade from providing basic voice with simple text messaging (SMS) to a wide variety of speech and data services. GSM voice quality is generally excellent, coverage in most countries extends to over 99% of the population, and common technology ensures automatic international roaming for voice and SMS for virtually all users around the globe (customer services asidel).

CDMA currently services more subscribers in North America than GSM. Its popularity among operators is due to its very efficient use of the radio frequency assigned to them.

The Future

2.5G systems are far from providing a truly global service. A European GSM subscriber cannot roam onto a North American CDMA network with the same handset. But, as will be discussed later, in terms of network infrastructure these systems are not radically different. it is the different ways in which the systems manipulate the radio spectrum that make them incompatible. If these systems could both adopt a compatible air interface, then there would be potential for a truly global cellular radio mobile network.

The future of mobile telecommunications, certainly from the operators’ point of view, is data. It is estimated that by 2006 data services will generate 50% of revenues, with networks offering audio and video streaming, real-time road GPS and MMS (Multimedia Messaging Service).

Operators in many regions of the word have reached a penetration level exceeding 75% of the population when it comes to voice coverage. Future users will not just talk, they will see, share and do.

How do wireless mobile phones work?

Mobile devices are radio telephones. The electrical telephone was invented in March 1876 by Alexander Graham Bell. Guglielmo Marconi first formally presented the radio in Italy in 1894. The design of both devices has remained fundamentally unchanged since the original.

Cellular networks around the world today employ a variety of different technologies, and even now there is no one true global standard. All of the technologies however do have many common features and the fundamental concepts behind cellular networks are common to all.

Cellular networks divide the area to be serviced into smaller cells, each cell being the area of coverage around a cell tower, or Base Transceiver Station.

A cellular network operator will be assigned a range of frequencies to use by the governing body in that country. In order to use this frequency allocation efficiently, it is subdivided into a number of smaller channels. Each cell operates on one of these channels, no two adjacent cells operating on the same frequency. This practice is known as Frequency Division Multiple Access (FDMA) and is common to all cellular systems.*

All mobile devices have special codes associated with them, which are used to identify the phone, the subscriber and the network operator. The information about the individual subscriber may be held on a removable memory card (SIM card, or Subscriber Identity Module), or could be hard-wired into the phone itself as is the case with most non-GSM networks.

When you first power on the mobile device it searches for an SID, or System Identity Code — a unique 5-digit identifier assigned to each network operator by the local governing body — on the control channel. If it cannot find any such ID it will display a “no service” message.

When the device finds an SID, it compares it to the SID that is programmed into it or the SIM. If they match then it knows it is on the home network.

Once the mobile device has detected the network, it then transmits a registration request. The network updates its location database with the new information and sends a message to the device via the control channel letting it know what frequencies to use for uplink and downlink. When the mobile network receives a call from the landline system destined for that device, it consults the location database to determine what cell it is in, and routes the call accordingly.

A” mobile stations have a 15-digit identifying number called an IMEI number (Individual Mobile Equipment Identifier), this can normally be found written on a sticker underneath the battery. When a mobile station registers on the network it broadcasts its ESN (Electronic Serial Number) which is verified by the EIR (Equipment Identity Register) and used by the HLR to track the device’s location.

As the device moves toward the edge of a cell, the base station notes that the signal strength is diminishing. Meanwhile, the base station in the cell which the device is moving toward (which is listening and measuring signal strength on all frequencies) registers the signal strength increasing. The two base stations coordinate with each other through the controller, and at some point, the device is sent a signal on the control channel telling it to change frequencies — to therefore move from one cell to another. This procedure is known as hand off.

Don’t concern yourself too much with the amount of acronyms l have thrown at you already, all will hopefully become clearer in a moment.

This is not strictly true, but bear with me at this stage. I will look in more detail at the technologies employed by different cellular systems later on.

Analogue versus digital

The first generation of cellular technologies relied on analogue systems to provide the air interface (the link between the mobile station and the nearest BTS).

Second generation (and subsequent) systems use digital systems in preference to their analogue counterparts. Digital systems use the same radio technology as analogue systems, but they use it in a different way.

Some people are understandably confused as to how a mobile network can be digital. Simply put, both analogue and digital systems use analogue radio waves to communicate between the wired network (the cell towers) and the mobile device. But digital systems use the analogue wave to transmit binary information in a series of is and Os, rather than an analogue sound wave. Read on…

Analogue signals have a tendency to lose their integrity because of the problem of “noise” or interference. The loss of this signal power is called attenuation. When an analogue signal passes through an amplifier to boost the signal, the noise that has been picked up along the way is amplified as well. The further a signal travels the more noise it will pick up. This is why long-distance calls sound much worse than local calls. Provided that the signal to noise ratio is low, then the interference is not a problem. However, one the signal to noise ratio gets too high, then the information cannot be interpreted. Data communications, which rely on high levels of quality, are therefore not well suited to analogue transmissions.

Digital signals are not as susceptible to this problem as they can only represent two values: 1 or 0, whereas analogue signals can represent a range of values.

Digital signals do not use amplifiers to boost signal strength, instead digital signals are regenerated. A device known as a repeater examines the incoming signal, determines whether each bit has a value of 1 or 0 and then passes the information to the output device where a perfect signal is generated.

Any noise that was picked up along the way is removed. Therefore the same quality of signal is maintained throughout.

How does analogue-to-digital conversion work?

This section will examine how an analogue signal can be converted to a digital data stream, what constitutes data, and how digital data can be transmitted over a cellular radio link.

The best way I can think of to explain how analogue to digital conversion works is to compare vinyl LPs and CDs.

Thomas Edison is credited with inventing the first means of recording sounds in 1877. Speaking into a “collecting horn” caused a needle attached to the horn to vibrate. The needle could be used to etch these vibrations into a suitable medium, such as a wax or tin cylinder. Rotating the cylinder and allowing the needle to transfer the vibrations back to the horn could play the sound back.

This is essentially the method still used today in the preparation of vinyl records, albeit vastly improved.

The variable-depth grooves etched into a record represent an analogue wave.Unfortunately this means of recording is low in fidelity — the noise caused by the etching itself is recorded, and after repeated playback the grooves become worn out.

The goal with digital media, such as the CD, is to produce a high-fidelity recording (it sounds the same as the original) with perfect reproduction (it sounds the same every time it is played back).

To accomplish this, the analogue wave is converted into a series of numbers, and the numbers are recorded instead of the wave. An analogue-to-digital converter does this conversion. During playback, the numbers are converted back into an analogue wave by a digital-to-analogue converter (DAC). This analogue wave is then amplified and passed to the speakers to produce the actual sound.

Provided the numbers held on the CD are not corrupted, the analogue wave produced will be the same every time.

To understand how analogue information can be converted to a digital signal, it is necessary to understand how digital signals represent data.

What is data?

User data can be anything — a text document, a spreadsheet etc. As far as a computer is concerned all data is a sequence of 1s and Us, or binary code.

The simplest unit of data that a computer deals with is a binary digit, or bit, and can have two values — 1 or 0.

A bit must be grouped together with other bits in order to represent information.

A byte is an 8-bit character. By using 8 bits a byte can hold any value from 1 to 256:

In the example shown above, a byte consisting of the bits 00010000 represents a value of 16.

A byte can therefore be used to represent 256 different characters. For every character typed into your computer there is a corresponding numeric value (decimal). The conversion is carried out by a code known as ASCII (American Standard Code for Information Interchange). The ASCII table contains 256 characters, each with a corresponding number value.

The first 128 characters are the alphabet (in both upper and lower case), numbers 0 to 9, and all punctuation. The second 128 characters comprise of graphical symbols (the second half of the ASCII set is rarely used in data communications). Depending on the country for which your computer operating system is configured, the ASCII set will vary to include the necessary country specific characters.

ASCII is only a standard used in the western world, as most far eastern ideographic character sets feature many more than 256 symbols.

So how is an analogue wave converted to digital information?

Below is a representation of an analogue wave. Assume that the ticks on the horizontal axis represent 1/1000ths of a second.

When the analogue wave is sampled, there are two variables involved — the sampling rate and the sampling precision. The sampling rate determines how many samples are taken a second. The sampling precision controls how many gradations are possible when taking the sample. This is also known as Quantization. The illustration below shows a sampling rate of 1,000 per second with a precision rate of 10:

The green rectangles represent samples. Every 1/1000th of a second the converter looks at the analogue wave and converts its value at that point to the closest number between 0 and 9. Using this sampling scheme the resulting blue line is created:

You can see that a lot of information has been lost using this scheme. A higher quality image would be gained by using a higher sampling rate with a higher level of gradation. In the case of CD sound, fidelity is an important goal so the sampling rate is 44,100 samples per second and the number of gradations is 65,536. At this level the output of the DAC so closely matches the original waveform that the sound is essentially “perfect” to most human ears.

This sampling rate produces a lot of data — an awful lot. It takes two bytes to represent each of the possible 65,536 gradations (as discussed above, each byte can represent any value between 0 and 256 [28:256], but 216=65,536). Two sound streams are being recorded (one for each of the speakers in the stereo pair). A CD can store up to 74 minutes of sound. This represents:

44,100 samples/channel/second x 2 bytes/sample x 2 channels x 74 minutes x 60 seconds/minute = 783,216,000 bytes

When an analogue voice signal is converted to digital information for transmission over a cellular radio link, the sampling rate required is not this high, and the range of frequencies used by human speech is not as broad, so the amount of data transmitted does not even begin to approach this amount.

Mobile Data

What does a modem do?

With conventional dial-up landline connections a computer requires a modem to be able to transmit data over telephone lines.

Landline telephone systems are optimized for speech. As the human voice only operates within a distinct frequency range, networks are optimized to transmit only these frequencies, often employing a passband filter to remove any other signals. The copper cables used in landline networks can comfortably transmit the frequencies used in human speech, but is not able to reliably transmit signals outside this range.

Data transmissions require a much wider frequency spectrum and therefore cannot pass through the landline telephone system without some form of conversion. This function is performed by the modem. The data signal is MOdulated prior to transmission by the sending modem, and then DEModulated by the receiving modem.

The amount of data a modem can transmit depends on the type of modulation scheme it employs.

Modem speed is often referred to in terms of the baud rate. Technically, the baud is the number of changes that can be made to a standing wave in one second. When a modem is working at 300 baud, this means that the basic carrier frequency has 300 cycles per second. Due to restrictions imposed by the physics of the wiring, a dial-up phone line can go up to 2400 cycles, a baud rate of 2400.

Real modem speed is measured in bits per second (bps). If each cycle is one bit, the fastest rate at which data can be transmitted is 2400 bps. However, by using different types of modulation, more than one bit can be transmitted per cycle.

How is digital data transmitted over a cellular radio network?

As discussed above, the first analogue networks did not support the transmission of data, partly due to the high quality of radio link required by data connections, but also because in the early 1980s there was little demand for such a service.

When networks are referred to a being digital, it is only the method of conveying the information that is digital — the signal is still sent via an analogue radio wave. The encoder converts the analogue information to be transmitted into a series of 1s and Os. These binary digits are then sent over the analogue carrier wave (the radio wave).

There are several methods of doing this, but all methods operate according to the same underlying principle. In essence, data is transmitted by altering a standing wave (the carrier wave) to convey either a value of 1 or 0.

This altering the characteristics of a carrier wave to bear information is known as modulation.

One form of modulation is Amplitude Modulation (AM). The amplitude of the wave (in mathematical terms, its displacement from the horizontal) is modified to represent two possible values: 1 or 0. This method of modulation is susceptible to interference, however.

An alternative to amplitude modulation is Frequency Modulation (FM), also known as Frequency-Shift Keying (FSK). FSK alternates the frequency of the wave, using one frequency to indicate a O and another to indicate a 1.

The information is recovered from the modulated carrier wave at the receiver by a process of demodulation.

Due to the limitation of the bandwidth allocated to cellular network operators, the amount of data that can be transmitted is very small. GSM networks can handle up to 9.6 Kbps and CDMA networks up to 14.4 Kbps.

The amount of data required to reproduce the human voice, however, exceeds this limitation.

The human ear can detect sounds in the range of a few hundred Hertz up to around 20 Hz, but most of the frequencies used for human speech are in the range between 300 Hz and 4000 Hz.

To filter out the frequencies not required for human speech, a source encoder is used. This filters out “redundant” bits from the digitized data stream.

To represent the human voice, the analogue signal needs to be sampled 8,000 times a second. This sampling rate is often referred to as the Nyqvist interval after the Swedish scientist who isolated this value as being the optimum required to reproduce human speech reliably. Each sample is quantized to one of 256 values, requiring 8 bits of data per sample. Therefore:

8,000 samples per second x 8 bits per sample = 64 Kbps

Because the only a specific range of frequencies is being sampled, the sampling precision is referred to as non-linear quantization.

Whilst a data rate of 64 Kbps is fine for the landline networks, where bandwidth is not too much of an issue and error rates are also not so consequential, the air interface used by cellular networks is a far more limited resource and is much more difficult to work with. Cellular networks therefore use modulation schemes that combine forward error correction and compression techniques. l will examine both these techniques in a moment.

Data transmission

Data transfer can be either serial or parallel. Parallel communications are quicker as concurrent paths are used simultaneously to transfer larger amounts of data. But parallel communications are only feasible over short distances. For long distance communications serial has to be used. Over serial connections data is sent one bit at a time. As a whole byte has to be received before the resulting value can be determined, computers use a buffer to store bits whilst it is waiting for the whole byte to be compiled.

Data transfer can be either synchronous or asynchronous. in order for two modems to reliably exchange information they must have a reliable clocking mechanism so that the sampling rate being employed is synchronized. Synchronous data transfer involves synchronizing the clocks on both the sending and the receiving modems before any data is transmitted. Synchronous communications are only necessary where large amounts of data need to be transferred on an almost constant basis.

Asynchronous data transfer involves adding data to the data stream to allow the modems to synchronize clocks. Bytes are sandwiched by “start” and “stop” bits. If you consider that error correction methods also add bits to the data stream, this represents a significant overhead and is the reason why data transfer is so slow — transferring a 100Kb file involves sending a lot more information than just the file itself.

All cellular networks transmit data asynchronously. In order to understand how modems communicate with each other, it is necessary to clarify a few terms:

DTE

Data Terminal Equipment (DTE) is the combination of a computer, a port, and application software that communicates with a second application — the remote application over a telephone line. The remote application is also a configuration of a computer, a port and an application that makes up the second DTE.

Most computers use a chip called a universal asynchronous receiver/transmitter (UART) to convert the computer’s synchronous parallel data from the processor into asynchronous serial data ready for transmission to the modem.

DCE

The Data Communications Equipment (DCE) is the modern. A DCE is also known as data circuit-terminating equipment, and its purpose is to link the DTE to the communication line.

Various asynchronous standards define the interface or signaling that goes on between the DTE and the DCE, and between two DCEs.

In asynchronous transmission, data is transmitted one character at a time. A start bit and one or more stop bits bracket each character. Asynchronous transmission is also called start-stop transmission. The asynchronous characters are not evenly spaced along the transmission medium. In gaps between characters, the line is idle; nothing is transmitted. The characters are transmitted independently with regard to timing signals.

The R8232 standard defines nine electrical circuits used for protocol handshaking. The following sections describe what these nine circuits do.

Carrier Detect (CD)

When the local modem detects a carrier signal from the remote modem, it raises the voltage level on this lead, indicating to the computer that a carrier signal is being received from the remote modem. This signal is also called the Received Line Signal Detector (RLSD).

The carrier signal is a tone or frequency. You can hear it usually just after you hear the originating modem dial. It is the sine wave signal that digital data from the two computers is mapped onto by the modem’s modulation function. Without a carrier there can be no connection. If this signal is dropped during a connection, your communication software will usually display a lost carrier error message. If a carrier is not detected at the start of a connection, the calling or originating modem displays no carrier.

Receive Data (RD)

When the local modem demodulates data received from the remote modem (DCE), the demodulated data is sent to the computer serial port by raising and lowering the voltage level on this lead.

Transmit Data (TD)

The computer sends data to the local modem by changing the voltage level on the Transmit Data circuit.

Data Terminal Ready (DTR)

The computer turns this circuit on when it is ready to be connected to the phone line. If the modem does not get this “ready” signal, no attempt to dial can occur and no commands from the computer will be received.

Signal Ground (SG)

Signal Ground (SG) provides a reference level or benchmark voltage for the other leads (circuits). Asynchronous signals (events, state changes) are sent between the serial port and the modem by voltage changes in circuits. For example, the computer sends or asserts DTR to the modem by changing the voltage on pin 4 to the voltage that indicates ON.

However, to make sense of this voltage change, a constant benchmark voltage is needed; this is the voltage level on pin 5 that is considered to be the zero voltage level.

Note, that there is another ground, called protective ground, which is like the third prong on a three-prong appliance plug. Protective ground is pin 1 in 25-pin connectors, and if it is present, SG is usually strapped or connected to protective ground. This pairing is not possible on a 9-pin connector because it does not have a protective ground lead.

Data Set Ready (DSR)

The modem turns this circuit on (asserts DSR to the computer) when it is ready and physically connected to the phone line. If the terminal does not get this ready signal, any attempt to dial will fail.

Request To Send (RTS) - The RTS circuit is turned on by the serial port to tell the modem that it actually has data queued for transmission. if the modem has CTS (Clear To Send) turned on, actual transmission will start. CTS off, signalled by the modem to the computer, tells the serial port that the modem temporarily cannot accept any data for transmission. When the asynchronous application writes a buffer destined for the remote application, this causes the serial port to raise RTS. RTS and CTS are used for flow control between DTE and DOE.

Clear To Send (CTS)

The CTS circuit is turned on by the modem to signal the serial port that the modem wants to write to the computer. If the serial port has RTS turned on (it responds with RTS), the modem will actually transfer data to the serial port. Data is accumulated byte by byte in a buffer by the serial port. When the buffer is full, it is transferred to an application buffer and is then ready to be filled again by the modem. If the serial port cannot accept data from the modem (DCE), it turns RTS off and the modem waits for RTS on. If RTS is off most of the time, it usually means the application is too slow. The modem is receiving from the remote DCE faster than the local application can read and process data.

Ring Indicator (RI)

The Rl circuit is turned on by the modem to signal the serial port that there is an inbound call. The line rings for one second, and then there is a four second pause. The RI circuit is turned on for each ring. It serves as a wake-up signal to the DTE. The DTE responds by asserting DTR to tell the modem that it is ready to be connected over the telephone company line. If carrier is detected, then data transmission starts.

Modem buffers

When the asynchronous application does a write, the bytes are queued in one or more buffers. The computer operating system then empties the application buffers by sending or writing the bytes out the serial port to a buffer in the modem. The asynchronous application in a single write can fill a large buffer; for example, 1024 bytes. Because the modem sends or transmits data one byte at a time over the line, the modem buffer can be overwhelmed. Application data in the

modern buffer can be overwritten and lost. When the application is reading data, the modem can transmit too fast, causing data in the application buffers to be overwritten. To keep this from happening, for data going in either direction, flow control is implemented.

Flow Control

When the modem’s buffer is so full that additional data written into the buffer causes data to be overwritten and lost, the modem has to signal the serial port to stop sending data. For inbound data, if the application can’t process the data fast enough (that is, it empties the modem buffer too slowly), and then the serial port has to signal the modem to stop sending. This handshaking or signaling is called flow control. Typically, flow control will be implemented in one of three ways: RTS/CTS or hardware flow control; XON/XOFF or software flow control; or ENQ/ACK or Enquire /Acknowledge flow control.

RTS/CTS or Hardware Flow Control

With RTS/CTS flow control, when the serial port wants to send data to the modem, it signals Request To Send (RTS) to the modem. If the modem can accept the data (that is, the buffer is not too full), it returns a Clear To Send (CTS) to the serial port. If the modem cannot accept the data, it turns off the CTS signal. On a 9-pin connector, RTS is signalled by raising or lowering the voltage levels on pin 7; CTS is signalled by raising or lowering the voltage on pin 8. For example, if the modem cannot receive any more data, it drops the voltage on pin 8, turning off CTS. Sometimes the application cannot process its buffers fast enough. To keep the modem from over-writing data, the application can tell the modem to drop RTS. The DOE (the modem) senses that RTS is dropped and stops filling the application buffer. When the application has emptied some of its buffers, it raises RTS again and the modem starts filling the buffer.

XON/XOFF or Software Flow Control

When a modem is receiving data from the local terminal or serial port too fast for the modem to process the data, the modem will send an ASCII 19 or Device Control 3 character to the serial port. This is the default XOFF character. It is sent on pin 3 (RD).

When the modem buffer is no longer full (that is, data has been transmitted to the remote modem), the modem sends the XON character — an ASCII 17 or Device Control 1 character to the serial port, and the serial port starts sending data again. The characters used for XON/XOFF can often be changed in the application and the modem if desired.

One problem with the XON/XOFF is that the ASCII characters sent to the serial port, could be characters embedded in a data file. The data is unintentionally interpreted as a control signal, causing transmission errors.

ENQ/ACK or Enquire / Acknowledge Flow Control

In ENQ/ACK flow control, when the serial port wants to send data, it transmits an ASCII enquire character. If the modem can accept data, it ACKs the serial port (transmits back the ACK, positive acknowledgement character). It if can’t accept the data (buffer full), the device or modem sends back a NAK, or negative acknowledgement, in response to the ENQ.

Error correction Cyclic Redundancy Check (CRC)

When the first modem has accumulated a buffer full of data for transmission, it applies a formula to the data block, calculating a value D. A second formula is applied to the data and the result is divided into the first value. The final result is a whole number quotient and a remainder.

Apply first formula to data. Result = D

Apply second formula to data. Result = G

Calculate D/G = Q + R

D is the result of applying the first formula to the data block. G is the value of the second formula. The remainder R is the cyclic redundancy check (CRC). It is usually 16 bits long and is appended to the end of the data block when the modem, which is not sending at that time, transmits.

When the receiving modern gets the data block, it goes through the same calculation. If there was a transmission error, the second CRC value will not equal the original one, and the receiving modem transmits a negative acknowledgement (NAK). The originating modem then retransmits the data block. If the two CRCs match, the receiving modem transmits a positive acknowledgement (ACK). The most commonly used formula, or polynomial, is called CRC-16.

Other methods

The simplest method of data compression is known as Run Length Encoding (RLE). RLE counts repeated characters or bit patterns, and then replaces the repeated characters with a representative bit pattern and a multiplier equal to the number of times the pattern or character is repeated.

The Lemple—Ziv—Welch (LZW) algorithm inspects the data for recurring sequences of information. LZW then builds a dictionary of repeated sequences. A pointer to the appropriate dictionary entry can now replace each repeated sequence in the data stream.

Huffman encoding looks at the data and counts which characters are most frequently used. These characters are assigned the shortest bit values. At the same time, the least frequently occurring characters are assigned the longest bit values.

Parity Bit

Bits are sometimes dropped because of noise. A parity check can be added to the data packet to recover from errors in which a single bit is dropped or added. Many communication software packages allow you to configure parity settings, or you may also be able to specify parity information through the serial port configuration options in an operating system. Parity, the number of stop bits, and the number of data bits, can all be set as serial port parameters.

if a parity bit is used, then an extra bit is added to each character to make the total number of 13 in the character either odd or even, depending on which parity type is used. If you configure your communication software to use parity bits, then there are seven data bits. If parity is not used, then there are eight data bits.

Compression

Compression works by examining the data to be transmitted and substituting commonly occurring bit sequences with a code that denotes that sequence. Provided that the receiving device supports the same compression protocol it will know to substitute that code with the actual bit sequence.

AMPS (Advanced Mobile Phone Service)

AMPS is the first generation analogue cellular technology still widely in use in North and South America today.

AT&T began development in 1960s and first commercially launched it in 1983 in Chicago. AMPS uses a range of frequencies between 824 MHz and 894 MHz for analogue cell phones. Due to the US government’s laws on monopolies — each major city in the US must be serviced by at least two network operators. These are referred to as A and B carriers.

Carriers A and B are each assigned 832 frequencies: 790 for voice and 42 for signaling (not to be confused with data). A pair of frequencies (one for transmit and one for receive) is used to create one channel. The frequencies used in analog voice channels are typically 30 kHz wide – 30 kHz was chosen as the standard size because it gives you voice quality comparable to a normal wired telephone.

The frequencies used for uplink are separated from those used for downlink by 45 MHz to keep them from interfering with each other. Each carrier therefore has 395 voice channels, as well as 21 data channels to use for housekeeping activities like registration and paging.

GSM (Global System for Mobile Communication)

The development of GSM began properly in 1982, when the European Conference of Posts and Telecommunications Administrations (CEPT), consisting then of the telecommunication administrations of twenty six nations, created a working group to develop a set of standards for a pan-European cellular network — the Groupe Spéciale Mobile (the origin of the acronym GSM).

The first GSM (Global System for Mobile Telecommunications, as the acronym now stands for) networks were not publicly launched until 1992. Before then the first generation of mobile telecommunications services (1G) were almost entirely analogue, unsecured and had little to no ability to handle data transmissions. Individual nations developed mobile networks independently of each other and roaming between countries with the same handset was impossible.

The first decision the Groupe Spéciale Mobile made was to adopt a digital system. Digital communications provide several benefits over analogue systems: security, data-capability and also devices can be physically smaller. It was originally intended that the digital mobile network would interface with the landline ISDN network, but limitations of the radio link meant that data rates of only 9.6kbps could be achieved compared to the 64kbps of the landline system.

GSM was originally only licensed for use in the 900 MHz frequency range. However, today GSM operates on 1800MHz and 1900MH2 due to local limitations in the allocation of available frequencies to mobile operators. Because if this, you may hear GSM networks being referred to as D08 or PCS networks.

DCS (Digital Communication Service) is identical to GSM but operates on 1800 MHz. The advantage of using the higher frequency means that mobile stations can use much lower power levels.

PCS (Personal Communications Services) is the umbrella term used to refer to networks operating at 1900 MHz in North America. The PCS initiative was launched when mobile operators realised that the older AMPS system was rapidly nearing saturation level and that an alternative was necessary. PCS technologies include D-AMPS, GSM and CDMA.

GSM Network Infrastructure

A GSM network is made up of a number of key components. The first of these is the subscriber’s mobile device. I use the term “device”, as it may be a mobile phone, PCMCIA radio card, cellular engine or any other device that incorporates a GSM transceiver and SIM card. In GSM terms this referred to as the terminal or “mobile station”.

The terminal contains an antenna that sends and receives radio signals to the nearest cell tower, or Base Transceiver Station (BTS). The radio coverage of an individual BTS is called a “cell”.

BTS are located every few miles, sometimes closer if the landscape prevents effective propagation of radio signals (or in areas of dense population). A country may have many thousands of BTS. Increasing or decreasing the transmitting power can vary the effective coverage area of a single BTS.

A handful of BTSs will be connected to a Base Station Controller (BSC). The BSC monitors the performance of individual BTSs and also enables subscribers to-pass from one cell to another (“handover”).

Multiple BSCs will be linked into the network’s Mobile Switching Centre (MSC). A country will only have several MSCs.

The MSC handles the routing of calls between GSM cells (if the call is made between two subscribers on the same network) and also to the terrestrial PSTN (Public Switched Telephone Network) and ISDN (Integrated Services Digital Network) networks.

The MSC that routes calls to the terrestrial network is known as the “gateway MSC”.

Located at each MSC is a component known as the VLR (Visitor Location Register). The VLR is responsible for keeping track of all subscribers and their cell locations within the geographical region of the nearest MSC. By knowing a subscriber’s location a GSM network operator can route incoming calls with the minimum impact on network resources.

The VLR is a dynamic database that has up to the minute records of where mobiles are, which mobiles are in conversation and which mobiles are banned for various reasons. When a mobile is switched off, the VLR typically removes that mobile from its database, thus keeping the number of active subscriber records to a minimum.

This information is also passed onto a larger central computer known as the HLR (Home Location Register).

The HLR is responsible for routing incoming calls from the terrestrial network to the required part of the GSM network.

The HLR also contains information on current services available to individual subscribers (data, fax, roaming, call forwarding etc).

In addition to the HLR, the central MSC also houses two computers, the EIR (Equipment Identity Register) and the AuC (Authentication Center).

Together these computers verify the identity of all subscribers by checking the information contained both within the mobile device and the Subscriber identity Module (SIM).

The Gateway MSC interfaces with the landline telephone system via an IWF, or inter-Working Function (also referred to as an IWU — Inter—Working Unit). This is essentially a bank of modems which converts the incoming digital data stream into the analogue tones used by the PSTN.

The entire network is divided into three subsystems:

1. Base Station Subsystem (BSS) — BTS and 880

2. Network Switching Subsystem (NSS) — MSC and HLR

3. Network Maintenance Subsystem (NMS) — monitors the performance of all

network components

GSM frequency usage

For GSM to succeed as a global standard, all member countries would ideally operate networks on the same frequency. However, despite the best efforts GSM still operates on three different frequencies:

900 / 1800 MHZ

1900 MHz (North America only)

Handsets operating on two or all of these frequencies are referred to as “dual- band” or “tri-band” devices.

These frequencies were chosen because they were available in almost all of the countries that wished to implement the standard. But they also serve practical considerations. A radio receiver needs to be as large as half the wavelength of the frequency it is designed to receive. Low frequency waves have long wavelengths. High frequency waves have short wavelengths, which means that the network’s radio interface does not have to be very large.

With the small amount of the frequency spectrum available, to accommodate the large number of mobile subscribers the networks developed some clever techniques to increase capacity.

A combination of three techniques is used today: “’ FDMA (Frequency Division Multiple Access)”

Networks that operate on 900 MHz actually use the frequency range 905 to 915 MHz for uplink, and 950 to 960 MHz for downlink. 2 operators, giving 5 MHz for each operator, share this 10 MHz “passband”.

This 5 MHz “bandwidth” is further sub-divided into 25 “carriers” of 200 KHz each. 24 carriers are available for subscribers while one carrier is reserved for network control signals.

Cellular Frequency reuse

The twenty four 200 KHz carriers are distributed among a number of abutted cells and then re-used over and over again.

This is possible as long as no two adjacent cells use the same carrier frequency. Using this technique only a small number of carriers can service a whole nation of subscribers.

TDMA (Time Division Multiple Access)

Each 200 KHz carrier is further divided into 8 “time-slots”, with one subscriber assigned to each time-slot.

Each subscriber on the carrier transmits for just 0.5 milliseconds before passing communications over to the next subscriber on the same carrier. This process is repeated in a cyclic pattern with each subscriber transmitting for an eighth of the time. Because each user is only able to transmit for the duration of the timeslot, voice and data traffic is buffered and transmitted in short bursts. The time slots are so small and the cyclic rate so high that the user perceives an uninterrupted communication channel.

One weakness with this system is that data has to be transmitted to preserve the individual data streams. That is to say, if one user has no data to transmit, “blank” data must be sent over the carrier to ensure that the other time slots do not fall out of sync.

This represents a further eightfold increase in the possible number of subscribers per cell.

Each cell has a finite number of subscribers it can service. This is based on the number of carriers multiplied by the number of time-slots per carrier (8).

The size of a cell can be multiplied by the increasing the output power of the BTS. In this way small-diameter cells in close proximity can service urban areas, whereas larger—diameter cells can service rural areas.

GSM Data Communications

GSM was the first mobile standard to allow for data and fax communication. Due to limitations in technology and available bandwidth, the speed of transmission was limited to 9600 bits per second per time slot allocated to the subscriber.

In terms of performance this equates to about 1.5 pages of A4 text being transmitted every second.

For fax communication 9600 bps is comparable to the performance of a landline fax machine, but for data transfer this is less than a fifth of the speed of a modern fixed line modem.

This speed limitation puts severe restrictions on the applications that can be used by a mobile subscriber.

The following applications will operate adequately at 9600 bps:

Text-based email

WAP

Instant messaging

Applications involving graphic or file transfer require a much higher data rate.

Making a data connection over GSM is similar to making a data connection via a landline.

1. A number is dialled and after about 20 to 30 seconds a circuit is established through the network to the destination.

2. While the circuit is open, the operator charges the subscriber for the duration of the call. This is referred to as a “circuit-switched” call.

Limitations

There are many weaknesses to this approach to data communications:

1. Subscribers have to wait up to 30 seconds to get connected to the Internet

2. Subscribers are billed for “dead” time when they are not necessarily transferring any data but the connection is still open.

3. While the subscriber is connected to the internet, but not necessarily sending or receiving data, that channel is not available to any other subscriber, effectively wasting the resource.

CDMA (Code Division Multiple Access)

i imagine that most of you reading this in the UK will not have come across CDMA before, as Europe has unilaterally adopted GSM as its cellular technology of choice. CDMA is not new, however. It has enjoyed widespread implementation in North America and services more subscribers there than GSM networks.

Code Division Multiple Access was first developed by the US military during the 1940s as a robust transmission system, which could withstand jamming attempts and was developed as a mobile telecommunications standard by Qualcomm in 1993.

CDMA is a spread spectrum technology. The term “spread spectrum” simply means that data is transmitted in small pieces over a number of frequencies within the entire available frequency range, rather than within a single defined channel. Digital Spread Spectrum (DSS) was developed during World War ll by the Americans as a robust communications standard resistant to deliberate or natural interference. It was initially used to direct torpedoes but has developed into a standard used in many different military and commercial applications, including cordless phones.

Bizarrely an American actress called Hedy Lamarr invented it. Lamarr and her arranger invented the frequency—hopping concept. Since Lamarr didn’t use her stage name on the patent, it took years for the story of patent 2,292,387 to surface. It never made her any money.

There are two approaches to spread spectrum: Direct Sequence Spread Spectrum (D888) and Frequency Hopping Spread Spectrum (FHSS).

DSSS works by splitting data up into a number of packets and sending them simultaneously over a number of different channels.

FHSS works by sending a short burst of data over one frequency, changing to another frequency, then sending another burst of data. The choice of the next frequency is random, so it is nearly impossible for someone to eavesdrop or jam the signal. Both sending and receiving devices need to agree on the frequency- hopping algorithm used before transmission. This mechanism allows for several different FHSS systems to work side by side without interfering with each other.

D888 is capable of much greater speeds than FHSS, but FHSS is less prone to interference than DSSS. CDMA uses FHSS technology.

Instead of dividing the frequency range available to the network operator into discrete channels and splicing users into that channel using a time—based multiplexing algorithm like TDMA systems, CDMA assigns each user an individual code and spreads their transmission over all of the available bandwidth.

One weakness of the TDMA system is that when a user has no data to transmit, “blank” data must be sent in order to prevent the other time slots from becoming out of sync.

CDMA makes a much more efficient use of the available bandwidth, and due to this improved efficiency allows for a higher number of subscribers per cell compared to GSM (as much as five times as many) and also enables data rates of up to 14.4 kbps.

By increasing the bandwidth available to a single channel, the harmful effects of interference (either deliberate or inadvertent) are reduced giving rise to high signal quality.

The choice of frequency is not entirely random — if it were the two parties involved in communication would have no hope of remaining in contact with each other. They employ what is known as a pseudo-random algorithm to decide what frequency-hopping pattern will be used for the duration of the communication. The algorithm is agreed upon during the initial channel setup negotiations.

The code in use may be different in each cell, so this negotiation needs to be conducted during the hand-off procedure each time the mobile device moves from cell to another.

Security is not the principle reason for adopting this system for use in public networks, although it is an added bonus. The main reason is the shortage of available frequency spectrum. Spread spectrum gives engineers a way to fit mobile devices into existing spectrum without jamming the devices already using it. Assume a phone is transmitting at 1 watt, but is hopping between dozens or hundreds of channels very rapidly. Other devices don’t “see” the phone because

it is transmitting for only a fraction of a second on any channel. Therefore, the average perceived power on any given channel is extremely low, and other devices using that channel don’t even notice it. The phone creates the equivalent of a low-power noise pattern across all of the channels it uses. Other devices deal with noise already, so the phone is essentially invisible to devices using specific channels.

This system has several other benefits:

increased network capacity

Less interference meaning higher audio quality

Increased reliability for data connections

Reduced power output meaning less interference caused to other devices

Reduced health risks

The main benefit to network operators, therefore, is increased capacity. Whereas with conventional FDMA (Frequency Division Multiple Access), networks re-use frequencies in a 7-cell pattern ensuring that no two abutting cells use the same frequency:

With CDMA each user is assigned a different instance of the carrier wave, a different phase of the wave:

Each cell therefore uses the entire passband.

The CDMA systems currently in use in the world today are referred to as cdmaOne systems — they are the first generation of CDMA systems (but remember that CDMA itself is still a 2G technology). HSCSD (High Speed Circuit-Switched Data)

Orange have implemented, in addition to a GPRS service, another 2.56 system. HSCSD offers data rates of up to 28.8 kbps by offering subscribers more than one timeslot.

By reducing the amount of error correction involved, the amount of redundant data transmitted is reduced so users get an effective data throughput of 14.4 kbps as opposed to 9.6 kbps.

By allocating subscribers two time slots, they can establish a connection to the Internet at 28.8 kbps.

Multi-slot operation

As incoming and outgoing data is transmitted on different frequencies, it is possible to operate different “multi-slot schemes”. This enables the operator to “weight” either the sending or receiving bandwidth to suit the subscriber’s application. For example, a 3-down 1-up scheme would suit web browsing as very little data is ever uploaded.

GPRS (General Packet Radio Service)

GPRS is a 2.5G technology implemented by GSM network operators to increase network capacity (to service the growth in the number of subscribers), to provide higher data rates for subscribers (to address the increased need for high-speed connections to the Internet and corporate networks) and also as an intermediate step on the path towards 3G systems.

GPRS uses the same physical network infrastructure and is designed to coexist with GSM. The air interface is still based on TDMA structure, with subscribers being allocated one or more “timeslots”.

A conventional GSM device is either on or off. When it is on, it is attached to the network and the HLR tracks the device to know which Base Station is it registered on.

When in a call, a conventional GSM device maintains a dedicated circuit from endpoint to endpoint. The mobile station is then the sole user of a timeslot. When no data is being sent, the network resources are still consumed.

GPRS is an extension of GSM. Instead of requiring a phone number to be dialed and a permanent circuit to be created until the user disconnects, GPRS is packet-based. When the mobile station registers on the network, it is assigned an IP address that enables data to be routed to it by other nodes on the network. The mobile station is then able to send and receive data almost immediately. This is referred to as an “always on” connection.

GPRS has a number of advantages over CSD dial-up Internet connections:

1. Unlike CSD, no 20 to 30 second delay is required to establish a connection. The GPRS terminal is able to exchange data with the network at a moment’s notice. This is referred to as being “always on”.

2. The network operator can monitor how many data “packets” are exchanged with the subscriber’s terminal, enabling a billing system based on the amount of data transferred rather than the length of a call.

3. As radio resources are only used when the subscriber is sending or receiving data, this more efficient use of the resources enables more subscribers to be serviced by a single cell.

4. The fourth benefit is a greater data transfer rate.

The mobile station still registers on the network in the same way as before, and the location of the terminal is still tracked by the HLR. GPRS does require that additional components be added to the network, however.

Each Base Station is upgraded with a Packet Control Unit (PCU) which determines whether the data being sent by the mobile station is voice traffic, CSD traffic or GPRS traffic, and then routes it accordingly.

GPRS data traffic is then sent to a Serving GPRS Support Node (SGSN) which works in a similar way to the MSC (Mobile Switching Center). Its main function is that of a router, examining the destination address of each packet and routing it according to the information held in its routing tables.

Data that is destined for a node that is not on the same network is routed to the Gateway GPRS Support Node (GGSN). The gateway between the mobile network and the wired Internet is also known as the Access Point Node (APN), which also records billing information.

Because GPRS uses Internet technology to identify nodes on the network and in the transmission of data, it makes it relatively simple to integrate with other Packet Data Networks (PDNs) such as the Internet and corporate networks.

Handset classes

GPRS mobile stations can fall into one of three categories.

Class A handsets will support simultaneous attach, activation, monitoring, invocation and data transfer on both GPRS and GSM modes. Calls can be made or received on both services. This is the most sophisticated class of device.

Class B handsets will be able to attach to, activate and monitor both GSM and GPRS services simultaneously but only be able to support data transfer on one at a time. Active GPRS “virtual circuits” are not cleared down when the GSM facilities are in use, but any attempt to contact the device’s assigned IP address will result in a busy or “held” indication.

For instance, if an incoming GSM call is answered, the GPRS connection is put on hold. This may cause problems for some applications, eg a file transfer may be aborted because the transfer protocol timeout expired.

Class C handsets will only be able to operate exclusively in one mode at a time. Manually changing from one mode to another involves detachment from the previously selected service.

GPRS Attach

When you activate your GPRS terminal, the nearest SGSN will assign the device an available iP address automatically. This process is known as the GPRS Attach. Once attached, GPRS connectivity is provided by a “context”. To be able to send and receive data, you will need to enter the address of the Access Point

Node, or GGSN. This is the APN which all of you will be familiar with and is usually a “friendly” text phrase.

The HLR will then track what SGSN the mobile is attached to, rather than what BTS it is physically communicating with, to be able to route incoming data packets correctly.

For devices capable of supporting simultaneous GSM and GPRS attach (Class A and Class B), the device may be registered on different cells for GSM and GPRS as the network sees fit.

PDP Context Activation

In order for the user to be able to transfer data, a Packet Data Protocol (PDP) Context must be activated in the MS, SGSN and GGSN. The user initiates this procedure, which is similar to logging on to the required destination network. The process is described below.

1. The user initiates the logging on process, using an application on the PC or MS.

2. The MS requests sufficient radio resources to support the Context Activation procedure.

3. Once the radio resources are allocated, the Mobile Station sends the Activate PDP context request to the SGSN. This signalling message includes key information about the user’s static IP address (if applicable), the QoS requested for this context, the APN of the external network to which connectivity is requested, the user’s identity and any necessary lP configuration parameters (e.g. for security reasons).

4. After receiving the Activate PDP context message, the SGSN checks the user’s subscription record to establish whether the request is valid.

5. If the request is valid, the SGSN sends a query containing the requested APN to the DNS server.

6. The DNS server uses the APN to determine the IP address of at least one GGSN that will provide the required connectivity to the external network. The GGSN IP address is returned to the SGSN.

7. The SGSN uses the GGSN IP address to request a connection tunnel to the GGSN.

8. Upon receiving this request the GGSN completes the establishment of the tunnel and returns an IP address to be conveyed to the MS. The GGSN associates the tunnel with the required external network connection.

Once this procedure is completed, a virtual connection is established between the MS and the GGSN. The GGSN also has an association between the tunnel and the physical interface to the external network. Data transfer can now take place between the MS and the external network.

GPRS Context Deactivation and Detach

GPRS provides two additional, independent, procedures that enable a PDP context to be deactivated and the MS to disassociate itself, or detach from the network. GPRS detach can be performed when:

- the MS is powered off

- the user wishes to detach from the GPRS network, but wants to remain connected to the GSM network for circuit switched voice services.

What happens to an incoming voice call during a GPRS data session?

GPRS Class B mobiles can be attached to both the GPRS and GSM networks, but they cannot transmit or receive on both simultaneously. If a mobile is in an active GPRS data session when an incoming voice call is detected, the user will be notified by an on—screen message, and then has the option to suspend the data session and accept the call, or continue with the data session and reject the call.

GPRS Coding Schemes

GPRS is specified on paper to achieve a maximum performance of up to 171 kbps. Realistically network performance will vary from anywhere between 9 to 50 kbps.

Network operators can choose to implement one of four different “coding schemes” to attain higher rates of data throughput.

in all data communications user data is broken down into smaller units called “packets”. A packet is comprised of the data itself (the “payload”), the address of the source machine and destination machine, and also “redundancy”, bits used for error correction to ensure that the packet has been received successfully.

With a CSD connection this redundancy can account for 50% of a data packet.

GPRS reduces the amount of redundancy used to increase the amount of user data that can be included in each packet.

Different coding schemes reduce this level of error correction to different degrees, thus making the connection less reliable, but increasing the data throughput.

The four schemes are described in the table below:

[table]

Due to the requirement for reliable connections, networks will typically implement either coding scheme 1 or 2.

Due to the allocation of radio resources by the network, GPRS speeds will fluctuate throughout the day. The more voice subscribers there are on the network, the fewer the resources available for GPRS data. This is due to the way in which networks use spare network capacity to accommodate GPRS traffic.

It must be remembered that whilst GPRS is suitable for “bursty” applications such as web browsing, it is not suited to all applications, such as video streaming, which require constant connections.

GPRS Multi-slot Operation

As with an HSCSD connection, GPRS is able to use more than one time slot to increase data throughput. The speed of each time slot will depend on the coding scheme used by the network.

As incoming and outgoing data is transmitted on different frequencies, it is possible to operate different multi—slot schemes. This enables the operator to “weight” either the sending or receiving bandwidth to suit the subscriber’s application. For example, a 3-down 1-up scheme would suit web browsing as very little data is ever uploaded.

CDPD (Cellular Digital Packet Data)

CDPD is an AMPS overlay in much the same way that GPRS is an extension to GSM.

Although AMPS is itself an analogue technology, CDPD itself is entirely digital, sharing frequency with the AMPS system but only transmitting in the blank spaces between AMPS voice calls.

Substantial amounts of data can be transmitted by hopping from one “gap” to another harnessing the otherwise unused channel capacity.

This channel hopping occurs completely transparently to the user.

CDPD data traffic was originally expected to be infrequent bursts of data typical of telemetry or credit-authorization applications. it quickly became apparent that during busy periods there were few if any channels available for CDPD to switch to and system became “blocked”. The carriers resolved this problem by offering dedicated channels for data traffic.

CDPD offers data rates of up to 19.2 Kbps. Error control overhead means that actual throughput of useful data can be up to 12 Kbps on a clean, lightly—loaded channel. Data transfer rates may be lower on congested networks with many voice or CDPD transmissions underway.

Like any data communications system, each CDPD data link has a maximum capacity it can support. The maximum 19.2 kbps throughput of CDPD limits the amount of data that can be sent over the channel. On average, if a user has an application that requires 5% of this maximum channel capacity, then one radio channel data link can support 20 users. On average, if a user has an application that requires 1% of this maximum channel capacity (not unusual, especially for light Internet connectivity), then a single radio channel data link can support 100 Users.

The nature of most user data applications is that the amount of data sent is small (a few frames) and the rate at which the data is sent is bursty (short periods of activity, followed by long periods of idle time). As a result, a CDPD radio channel data link can support a large number of users at one time. The maximum number of users that can be supported on a single radio channel data link depends on the nature of the data traffic that the users’ applications send.

Like wired Ethernet networks, CDPD is a “contention-based” system. It uses DSMA—CD (digital sense, multiple access, collision detect), while Ethernet uses CSMA—CD (carrier sense, multiple access, collision detect). in both systems, when a device has data to send it senses the transmit medium to determine if it is currently busy. If not, it will send its data. If it is, it will wait a random interval before trying again.

Carrier Sense Multiple Access / Collision Detection (CSMA/CD)

Let‘s represent our radio channel as a dinner table, and let several people engaged in polite conversation at the table represent the mobile devices trying to use it. The term Multiple Access means that when one mobile station transmits, all the stations on the medium hear the transmission, just as when one person at the table talks, everyone present is able to hear him or her.

Now let’s imagine that you are at the table and you have something you would like to say. At the moment, however, I am talking. Since this is a polite conversation, rather than immediately speak up and interrupt, you would wait until I finished talking before making your statement. This is the same concept described in the Ethernet protocol as Carrier Sense. Before a station transmits, it “listens” to the medium to determine if another station is transmitting. If the medium is quiet, the station recognizes that this is an appropriate time to transmit.

Carrier Sense Multiple Access gives us a good start in regulating our conversation, but there is one scenario we still need to address. Let’s go back to our dinner table analogy and imagine that there is a momentary lull in the conversation. You and I both have something we would like to add, and we both “sense the carrier” based on the silence, so we begin speaking at approximately the same time. In network terminology, a collision occurs when we both spoke at once.

In our conversation, we can handle this situation gracefully. We both hear the other speak at the same time we are speaking, so we can stop to give the other person a chance to go on. Mobile stations also listen to the medium while they transmit to ensure that they are the only station transmitting at that time. If the stations hear their own transmission returning in a garbled form, as would happen if some other station had begun to transmit its own message at the same time, then they know that a collision occurred. A single Ethernet segment is sometimes called a collision domain because no two stations on the segment can transmit at the same time without causing a collision. When stations detect a collision, they cease transmission, wait a random amount of time, and attempt to transmit when they again detect silence on the medium.

The random pause and retry is an important part of the protocol. If two stations collide when transmitting once, then both will need to transmit again. At the next appropriate chance to transmit, both stations involved with the previous collision will have data ready to transmit. If they transmitted again at the first opportunity, they would most likely collide again and again indefinitely. Instead, the random delay makes it unlikely that any two stations will collide more than a few times in a row.

Network infrastructure

To implement a CDPD service, AMPS operators need to upgrade the Base Stations to Mobile Data Base Stations (MDBS). These perform the same function as the Packet Control Units in a GPRS system: they examine the data and determine where it needs to be sent. A series of Mobile Data intermediate Stations (MD-IS) control a number of base stations and are responsible for managing access to the radio channel and channel hopping. They are also responsible for forward error correction.

CDPD is a packet-based system and uses Internet addressing schemes. As with GPRS, when the mobile station is activated, it will “attach” to the network and assigned an IP address by the network.

Introduction to 3G

Third Generation (3G) systems represent the future of mobile communications. Conservative estimates tentatively suggest that 3G services represent a cumulative revenue potential of one trillion dollars for mobile services providers between now and 2010.

A total of 2.25 billion mobile subscribers is forecast for 2010, of which more than 600 million will be 3G subscribers.

The goal of SG is to provide a truly global, compatible, cellular radio network. All of the current disparate technologies will adopt a common air interface providing greater network capacity, higher data rates, information-based services and global roaming.

This section will examine the services that operators are likely to offer and how they will be billed. I will look at the technologies that 3G will employ to attain these higher data rates, and also the steps that operators will need to take to evolve from the current 2.5G systems.

3G Services

3G services are likely to fall into one of three business models: the Access- focused, Portal-focused and Mobile Specialised Services approaches.

The six service categories that represent the major areas of demand for 3G-enabled services over the next 10 years can defined as follows:

* Customized Infotainment

* Multimedia Messaging Service

* Mobile Intranet / Extranet Access

* Mobile Internet Access

* Location-based Services

* Rich Voice (simple and enhanced voice)

The figure below shows how the UMTS Forum has defined SG-based mobile service categories:

[image]

By 2010, 28 per cent of the world’s 2.25 billion mobile cellular subscribers will be 3G subscribers. Figure 2 shows the worldwide revenues forecast from all 3G services for the 2001-2010 timeframe

Forecasts predict that total service provider-retained revenues for 3G services in 2010 will reach US$322 billion. Of those revenues, 66 percent will come from 3G-enabled data services. The cumulative revenue potential for mobile services providers between now and 2010 is over one trillion dollars. The consumer segment will contribute about 65 percent — a true mass-market success.

Customized Infotainment is the earliest and single largest revenue opportunity by virtue of its low cost and mass-market appeal, contributing US$86 billion in 2010.

As services mature, prices will decline. Services like email and Web browsing are likely to have little or no direct revenue potential; as users will expect them to be included as part of their service package at no additional charge. Their benefit comes mainly from their role as drivers of traffic. Much of the additional revenue generated by 3G services and applications will come through increased usage rather than through new sources of revenue.

3G is about services, not technology. Services will be interesting, easy to use, reliable, affordable (with transparent charging) and ubiquitous. Customers do not buy “bytes”, bit rates or technology, only content and services which are interesting on an individual/personalized basis, and which are affordable by “adding value” to lifestyles. Users will expect seamless anytime/anywhere access to voice, data, the Internet and multimedia services on a global basis.

Revenue is likely to be shared by the mobile operator and third party application developers and content providers. There are three likely business models:

1.) The Access-focused services provider offers mobile and lP network access. Users are “Internet-experienced” and prefer to browse directly to the sites they want. Likely service categories offered are Mobile Internet Access (consumer) and Mobile Intranet / Extranet Access (business). Revenue comes only from ISP subscription and airtime.

2.) The Portal-focused services provider offers access to the mobile IP networks and selected partner content, all via a mobile portal. Users are subscribers to voice mobile services, and are not expert Internet users. They appreciate easy access to content that is tailored to their preferences and interests. Revenue sources include subscription, airtime, transaction fees and advertising. The likely service category offered is Customized Infotainment (consumer).

3.) The Mobile Specialized Services services provider offers specialized service— sets targeted to a specific market. Revenue sources include airtime, messages, subscription, advertising, transaction and messaging fees. Likely service categories offered are Multimedia Messaging, aimed at a particular demographic group, and Location-based Services.

Positioning 3G as the “Internet made mobile” sends the wrong message to the market and paints an incomplete picture of 3G service potential. The advantages of mobility, personalisation, location capability and other features of 3G services rather than speed should be highlighted. Customer expectations should be managed; current fixed Internet subscribers will expect 3G services providers to offer a full, high-speed, low-cost Internet experience. 3G will replace some high-speed fixed Internet access, but will never be a complete substitute.

What is a mobile portal?

Portals will be critical to the end user’s experience of 3G services. Although superficially similar to the familiar portals of today’s fixed Internet, 3G mobile portals exhibit significant differences in character due to the additional challenges of content optimization for small form factor devices, and the necessity of delivering that content to the mobile user.

Defined as “an entry point to a wealth of information and value added services”, a portal is Internet/Intranet based with a browser-based interface, and can be personalised, delivering content according to a device‘s characteristics and a user’s needs.

Content and services provided by these portals can be considered as falling into six main categories: '

* Communications and community — email, calendar and chat

* information — news, weather, directories

* Lifestyle — listings of events, restaurants, movies and games

* Travel — hotel listings, direction assistance and timetables

* Transaction -— banking, stock trading, purchasing and auctions

* Other — information about personalization, location- based services, device type and advertising

There are five likely types of 3G mobile portal:

* Mobile Intranet / Extranet portal — a 3G portal that provides secure mobile access to corporate LANs, VPNs and the Internet. Typical services include corporate email, calendar, training and customer relationship management tools

* Customized Infotainment portal — a 3G portal that provides device— independent access to personalized content anytime anywhere. Typical services include streaming music, short film/video clips and m—commerce applications.

* Multimedia Messaging Services portal — a 3G portal that offers non-real—time, multimedia message access allowing the provision of third party content. Examples of typical services include multimedia postcards, video clips and movie trailers.

* Mobile Internet portal — a 3G portal that offers mobile access to content services with near-wireline transmission quality and functionality. Typical services include browsing, gaming and m-business.

* Location—based Services portal — a business and consumer 3G portal that enables users to find other people, vehicles, resources, services or machines. It also enables others to find users, as well as enabling users to identify their own location via terminal or vehicle identification. Typical services include emergency services, asset tracking, navigation and localized shopping information.

Implications of 3G

Insufficient battery life and power consumption issues will impede the ability of 3G terminals to deliver the full potential of mobile multimedia services in the short term. Research needs to be intensified into the area of battery technology.

A number of other significant technical issues will impact on the delivery and market acceptance of BG portal services. Perhaps the most contentious of these is security, where a current lack of privacy policies and open-standard solutions, coupled with uncertainty over strategies for dealing with mobile junk mail (spam) and viruses, presents several major challenges to industry.

Another area where there is a current lack of clear industry coordination is billing, charging and payment: while the financial industry and manufacturers are working to integrate micropayment technology into mobile devices, uncertainty remains over which billing systems will be accepted by end users and timescales to bring solutions to market. This issue will be of even greater relevance to 3G due to the large amounts of network traffic that will be generated by multimedia messaging and file transfers.

Demand for 36 mobile data services is real. Users have consistently demonstrated strong interest in trying new services that combine mobility with content and personalisation. The strong growth in SMS has clearly demonstrated market acceptance for messaging which has now evolved into Multimedia Messaging.

Billing

The move to 3G demands a fundamental shift in the manner in which mobile services and applications are billed, and will require a revolutionary transition from the operators’ perspective. Issues such as who supplies what to whom, and the fragmented segmentation of next-generation end-users will also mark a significant departure from the traditional mobile billing model. Billing systems that evolve to support 3G services will have to cater for a wide range of events and services, including revenue sharing and the apportionment of revenues to third parties.

The ability to bundle services will help to promote service personalization by allowing customers to tailor packages and encourage experimentation, for example, providing a movie package that allows, say, four films per month, a daily personalized news service and local cinema information. A premium would then be charged on top of the fixed fee for access to the latest film releases. Alternatively, the number of pages used or service duration on an ongoing basis could dictate fees. Again, the consumer chooses the service package to suit personal preferences and lifestyle.

Initial experience with packet-based services shows there are three basic components to the billing model:

* A fixed monthly data services subscription

* A fixed service or content area subscription

* A variable data traffic fee

EDGE (Enhanced Data rates for Global Evolution)

EDGE is a third generation (3G) system which proposes to offer a maximum data throughput of 473.6 Kbps (but as with all systems this is an absolute maximum and will not be attainable in a commercial system).

EDGE manages to offer such high data rates by using complex modulation techniques and a greater frequency bandwidth than current cellular radio systems.

The main benefit to EDGE, however, is that it can be upgraded to from all of the current 2.5G systems, offering the possibility of a single compatible global mobile network, comprised of individual network operators.

As is to be expected with such an area of research, there are a lot of acronyms in use — many different countries using different acronyms to refer to the same thing.

The theoretical global mobile network is referred to as the Universal Terrestrial Radio Access Network (UTRAN). All 3G technologies and concepts fall under the umbrella term UMTS (Universal Mobile Telecommunications System).

EDGE is also sometimes referred as EGPRS (Extended GPRS).

The principles of data transmission — a refresher

As you will remember from the introduction, data is sent over an analogue radio wave by altering the characteristics of that wave for a brief instant in such a way so that the receiving device can interpret from that change a value of either 1 or 0.

This technique is known as modulation. The most commonly used forms of modulation involve altering either the wave’s amplitude or frequency.

The more 1’s and 0’s that can be interpreted from a single change will mean a higher data rate.

Modem speed is referred to in terms of baud rate. Technically a baud rate is the number of voltage or frequency changes the device can complete in a second. When a modem is said to be operating at 300 baud, this means that the carrier wave has 300 cycles per second.

Real modem speed, however, is measured in bits per second. If each cycle (frequency change) can convey one bit (a value of either 1 or 0), then a modem operating at 300 baud can transmit 300 bits per second.

However, if a modulation technique is used that sends two bits per baud, then the modem will be able to transmit 600 bits per second.

Landline networks use a variant of amplitude modulation called Quadrature Amplitude Modulation (QAM) which sends two bits per baud. As all bits can only have 2 possible values, it follows that 2 bit groups can only have 4 possible values:

00

01

10

11

QAM works by altering the amplitude of the carrier wave by one of 4 degrees, each degree corresponding to one of the four possible values:

However, due to the susceptibility of radio waves to interference, this method is not suited to cellular radio systems. GSM uses a technique known as Gaussian Minimum Shift Keying (GMSK) which combines frequency modulation with encryption technology.

EDGE uses a different modulation technique that alters the phase of the wave. This is known as Phase—Shift Keying (PSK).

Waves have relative starting points, or phases. Two identical waves, which have different starting points, are said to be out of phase. If you’re interested in Hi-Fi you may have spent a considerable amount of time ensuring that your speakers are positioned so that the stereo image is focused on your normal listening position, ie to ensure that when the two sound waves from the speakers converge at the listening point they are in phase ….. or maybe it’s just me!

This change in phase can be used to send data. If you treat one wave as having a value of 0, then any wave that is out of phase can represent a value of 1.

EDGE uses a 4 phase shift pattern. This means that as QAM sends 2 bits per baud by using 4 waves of differing amplitude, so is EDGE able to send 2 bits per baud by using 4 waves of differing phases:

If the shift pattern is increased to 8 then 3 bits will be sent per baud and the rate of data throughput will be correspondingly higher. However, due to the reliance of EDGE on signal quality, the level of error correction necessary to guarantee an acceptable level of reliability means the net data throughput will not be a direct 3 to 1 ratio. EDGE has the ability to tailor the coding scheme used depending on the quality of the radio signal. In poor radio conditions, robust coding schemes are selected, whereas in good radio conditions less robust coding schemes are employed.

Network infrastructure

In terms of network infrastructure, little needs to be changed to upgrade an existing GSM/GPRS system to EDGE. The principle difference is that the air interface is changed. Instead of the FDMA / TDMA system, CDMA is employed instead. As discussed above, CDMA uses the available spectrum very efficiently and so can offer higher capacity and higher rates of data throughput than GSM. Using this air interface in conjunction with improved modulation techniques means that data rates will be increased many-fold.

EDGE will use a wider frequency range than is currently allocated to GSM/GPRS network operators. This additional frequency was auctioned in European countries not so long ago — you may recall the press coverage. The vast sums paid by network operators for this frequency almost bankrupted some of them, and returns on this investment will be slow in materializing. Due to the incompatibility of CDMA with existing GSM systems, new handsets will need to be developed and the two systems will need to be run side-by-side for some time.

This use of CDMA technology over a larger frequency range is referred to as Wideband CDMA (W-CDMA).

Globalization

The key to creating a globally-compatible internetwork of mobile cellular networks is the adoption of W-CDMA as a common air interface.

GSM is already the leading cellular standard worldwide and the evolution from cdmaOne networks to W-CDMA is not a difficult one. If all GSM, cdmaOne and

AMPS networks adopt W-CDMA as their air interface then a truly global network will be possible.

WAP — Wireless Application Protocol

http://www.wapforum.org

You may remember adverts on television urging viewers to “surf the net, surf the BT Cellnet.”

The service being advertised was WAP. WAP stands for Wireless Application Protocol and enables WAP-capable handsets to retrieve and display web pages in a micro browser.

WAP was first launched in 1999. Whilst the benefits and possibilities offered by WAP are many, due to the relatively arduous setup procedure, the slow connection speed and the limitations of phone displays, uptake was slow.

Before GPRS was launched, WAP connections were expensive as users would often forget that they were paying for a call once connected and would leave their connections open.

With the advent of GPRS and billing based on data transfer rather than call duration, WAP is coming into its own. OTA (Over The Air) provisioning tools such as Mouse2Mobile make the process of setting up the connection painless and even completely transparent to the user.

The companies who have joined the WAP Forum, companies such as Nokia and Ericsson, indicate the seriousness with which the technology is being taken, and it is not an exaggeration to say that all phones manufactured today contain a micro browser and are therefore WAP-capable. The adoption of the standard by all major handset manufacturers allows manufacturers, network operators, content providers and application developers to offer compatible products and services that work across varying types of digital services and networks.

The mobile phone is evolving into much more than a wireless telephone; it is becoming a portable data terminal. However, whilst WAP delivers web content to a subscriber’s mobile device, due to the limitations of phone displays and processing power, it is important to think of WAP in realistic terms. WAP uses Internet technology to deliver content, but in terms of user experience it is far from the multimedia extravaganza that the Internet is rapidly becoming.

Let us look briefly at typical Internet architecture:

When a user enters the address of a web page in their browser, an HTTP (HyperText Transfer Protocol) GET request is sent to the appropriate server, which delivers the HTML page itself to the user, as well as any graphics, videos or sound files that the HTML code links to.

WAP is an air link-independent protocol, meaning it can use CSD, HSCSD, GPRS or even SMS bearer services. There are a number of network components a network operator must install, however, to provide a WAP service.

The most significant component in the WAP solution is the WAP gateway. The WAP gateway is in essence the route through which all data must pass on its route from the Internet to the mobile phone. The gateway effectively condenses the protocols and languages of the Internet into the more condensed and efficient protocols used in WAP.

For example, the transport protocol used by the Internet is TCP. TCP offers the significant benefit of being able to check for errors during data delivery and can retransmit data should it get lost en route.

This constant signaling back and forth acknowledging the receipt of data constitutes a significant amount of data that the low bandwidth-capability of GSM cannot afford. Hence the equivalent protocol in the WAP environment is WDP.

Another significant thing that the WAP gateway does is to condense the information that is passing through it, in order to optimize the connection speed.

Whereas conventional web pages are written in HTML (HyperText Markup Language), WAP pages are written in their own text-based language called WML (Wireless Markup Language). However, being text based it is not efficient. The WAP gateway also compiles the text code into binary information, making it considerably smaller and therefore faster.

WML can only display text, bullets and simple black and white graphics. Although for many applications much more is not required as a great deal of information can be portrayed with just this simple level of display flexibility.

Security for WAP

With the ever-increasing reliance on the secure transmission of data over the Internet, the WAP standard was revised to include support for encryption and public key technology.

Version 1.2.1 of the WAP standard includes support for secure data transfer for applications such as WAP banking and online shopping.

Based on SSL technology, WAP uses WTLS (Wireless Transport Layer Security) and is supported in nearly all handsets today.

I don’t intend to go into too much detail about how encryption works, see my guide to encryption, authentication and digital certificates for more information on this subject.

Briefly, to ensure that data being transmitted over the Internet cannot be intercepted; it is encrypted using a key. Only someone in possession of the same key can decrypt the information.

To ensure that the encrypted data is coming from a trusted source, the encrypted information is then digitally signed using a unique identifier which is verified as being authentic by a third party certificate authority.

There are three modes of operation of WTLS, corresponding to three ascending levels of authentication. Class | WTLS includes no authentication, Class ll WTLS includes authentication of the server to the client, and Class III includes mutual authentication, ie authentication of both the server and the client to each other.

The Gap in WAP

The bridging of two secure connections at the WAP Gateway, where the WAP Gateway is run by the network operator, has been termed “the gap in WAP” by some media, since the data is decrypted for a brief period at the WAP Gateway.

This can be addressed if the content originator operates its own gateway, so that WTLS-encrypted packets can be routed through or around the wireless carrier’s gateway. If this is not the case then another solution is application-level security.

Application-level security

The data created by an application can be encrypted before it is packaged for transmission over the Internet and further encrypted.

WML provides support for application—level security via the use of its SignText function, which allows the creation of digital certificates at the application level.

Introduction to Computer Basics

To compute is to ascertain the value of a number by calculation. The earliest form of the computer is the abacus which was first invented by the Chinese over 2500 years ago and is still in use today.

The evolution of the computer

The computer has evolved from very humble beginnings to the power processing engines of today.

The first mechanical computer was the analytical engine, conceived and partially constructed by Charles Babbage in London between 1822 and 1871. It was designed to receive instructions from punched cards, make calculations with the aid of a memory bank, and print out solutions to mathematical problems. However the thousands of moving parts required for its construction was beyond the technical expertise of the day.

The first electrically-driven computer designed expressly for data processing was patented on January 8, 1889, by Dr. Herman Hollerith of New York. The prototype model was built for the U. S. Census Bureau and computed results in the 1890 census.

Using punched cards containing information submitted by respondents to the census questionnaire, the Hollerith machine made instant tabulations from electrical impulses actuated by each hole. It then printed out the processed data on tape. Dr. Hollerith left the Census Bureau in 1896 to establish the Tabulating Machine Company to manufacture and sell his equipment. The company eventually became IBM.

The first modern digital computer, the ABC (Atanasoff—Berry Computer), was built in a basement on the Iowa State University campus in Ames, Iowa, between 1939 and 1942. The development team was led by John Atanasoff, professor of physics and mathematics, and Clifford Berry, a graduate student. This machine included many features still in use today: binary arithmetic, parallel processing, regenerative memory, separate memory, and computer functions. When completed, it weighed 750 pounds and could store 3000 bits (0.4 KB) of data.

The 1960s and 1970s heralded the era of the mainframe computer using technology from the original ABC computer.

The first true personal computer with its own operating system arrived on the scene in 1977 in the form of the Apple II.

Microsoft released Windows 3.11 in 1992 and the rest is history.

Computer communications

Telegraphs and early radio communication used codes for transmissions. The most common, Morse code (named after its creator, Samuel F. B. Morse), is based on assigning a series of pulses to represent each letter of the alphabet. These pulses are sent over a wire in a series. The operator on the receiving end converts the code back into letters and words.

Morse notation uses two values - a dot or a dash. By stringing these values together the telegraph operator can form letters and hence entire messages.

This is a form of binary notation. Computers also use binary notation, but rather than deal with dots and dashes they deal with 13 and Os.

Computers are essentially very complicated switch boxes, each switch having only two possible states — on or off. A switch that is on has value of 1, and a switch that is off has a value of 0.

All information processed by a computer is a string of 1s and Us.

The smallest piece of information used by a computer is a bit — a single 1 or O.

A byte is a group of 8 bits. A byte is required to signify one character of information and can represent up to 256 different characters.

Parallel and Serial communications

The wires in a computer only allow one unit of data to pass through at a time, much like a tunnel that only allows one person to walk down it at a time.

To speed things up we can add more wires to allow more data to be sent at the same time. In early PCs eight wires are often together to allow a byte to transmitted. These eight parallel wires are called a bus.

Buses in modern computers are normally either 32 or 64-bits wide allowing for the simultaneous transmission of more data.

There are several different types of bus on a computer. The term bus refers simply to the wires connecting the different parts of a computer to each other. A modern computer has several different buses which may different in width.

An overview of computing

Despite the apparent complexity of computers, all aspects of a PCs functionality and its components fall into one or sometimes two categories:

Input

Processing

Output

[image]

Input refers to any device that enters information into the computer from the outside world, be it a keyboard, mouse, floppy disk, CD, DVD, scanner or modem for example.

Processing refers to the manipulation of the inputted data by the computer.

Output refers to the presentation of the manipulated data, be it a monitor, printer, CD—R, modem or sound card, for example.

Let’s look at the principle components of a modern PC

Microprocessor

The microprocessor is an integrated circuit that contains the CPU on a single chip.

External Data Bus

I explained above that all data moves around a PC via a bus. The external data bus (also referred to as simply the data bus) is the primary route for data in a PC. All data-handling devices are connected to it so that any data placed on the bus is available to all devices connected to the PC.

Motherboard

The motherboard defines the computer’s limits of speed, memory, and expandability. A computer needs more than just a CPU and memory. To accept input from the user, it needs devices, such as a keyboard and a mouse. it also needs output devices, like monitors and sound cards, to cope with the powerful graphics and sound capabilities of the programs available today. A computer also needs “permanent” storage devices, such as floppy disk drives and hard disk drives, to store data when it is turned off. It is the function of the motherboard to provide the connectivity for all these devices.

There are two main types of motherboard — AT and ATX, the main difference between them being the type of power supply used.

CPU

The CPU is the part of a computer in which arithmetic and logical operations are performed and instructions are decoded and executed. The CPU controls the operation of the computer. Early PCs used several chips to handle the task. Some functions are still handled by support chips, which are often referred to collectively as a “chip set.” The image below shows an Intel Pentium CPU:

[image]

The CPU handles two operations, which used to be performed by two separate chips: the Control Unit (CU) and the Arithmetic Logic Unit (ALU). The ALU handles the mathematical computations and the CU controls the flow of data.

The CPU uses Registers as temporary memory to store data while it is being manipulated. A register is a row of switches which are set to either on or off. The wider the register the higher the performance of the PC.

The chip set is often a collection of chips that come pre-soldered to the motherboard (the image above shows a VIA chipset). When buying a processor it is important to check that it will be compatible with the chipset of the motherboard. A typical chipset will consist of:

Bus controller

Memory controller

Data and address buffer

Peripheral controller

The BIOS

In addition to the chipset you will also find another chip pre-connected to the motherboard — the ROM BIOS. A ROM BIOS chip contains data that specifies the characteristics of hardware devices, such as memory and hard disk and floppy disk drives, so the system can properly access them.

ROM (read-only memory) is a type of memory that stores data even when the main computer power is off. This is necessary so that the system can access the data it needs to start up. When stored in ROM, information that is required to start and run the computer cannot be lost or changed.

BIOS chips can either be standalone and contain static information, or they can be used in conjunction with CMOS chips which store user—definable settings which the BIOS accesses. CMOS stands for Complementary Metal Oxide Semiconductor and gets its name from the way it is manufactured rather than from the information it holds.

The information contained in a CMOS chip will depend on the manufacturer. Typically, CMOS contains at least the following information:

Floppy disk and hard disk drive types

CPU

RAM size

Date and time

Serial and parallel port information

Plug and Play information

Power Saving settings

The BIOS is accessible when the PC first boots — you will normally see a line on the bottom of the screen prompting you to press F2 to enter Setup, or words to that effect. A typical BIOS screen looks something like this:

[image]

The Clock

Timing is essential in PC operations. Without some means of synchronization, chaos would ensue. Timing allows the electronic devices in the computer to coordinate and execute all internal commands in the proper order.

Timing is achieved by placing a special conductor in the CPU and pulsing it with voltage. Each pulse of voltage received by this conductor is called a “clock cycle.” A” the switching activity in the computer occurs while the clock is sending a pulse

The clock speed is measured in megahertz (MHz) and indicates how many commands can be completed in two cycles. The process of adding two numbers together would take about four commands (eight clock cycles). A computer running at 450 MHz can do about 44 million simple calculations per second.

Memory

The ability of the CPU to store data is very limited. To compensate additional chips are installed with the sole purpose of storing the data that the CPU may require. These chips are called random access memory (RAM). The term random access is used because the CPU can place or retrieve bytes of information in or from any RAM location at any time.

Address Bus

The contents of RAM is constantly changing as the CPU uses portions of it to hold data, manipulate it, then store the result of the calculation. It is vitally important that the CPU knows what memory locations are being used by what process and what memory is free to use. RAM is controlled by the Memory Controller Chip which passes requests to and from the CPU via the Address Bus.

[image]

How Microprocessors work

Let’s look at a simple task: adding two numbers such as 2 and 2 together and obtaining their sum (2 + 2 = 4). The CPU can do math problems very quickly, but it requires several very quick steps to do it. Knowing how a CPU performs a simple task will help you understand how developments in PC design have improved PC performance.

When the user pushes a number key (in a program like Calculator, which can add numbers), the keystroke causes the microprocessor’s prefetch unit to ask for instructions on what to do with the new data. The data is sent through the address bus to the PC’s RAM and is placed in the instruction cache, with a reference code (let’s call it 2 = a).

The prefetch unit obtains a copy of the code and sends it to the decode unit. There it is translated into a string of binary code and routed to the control unit and the data cache to tell them what to do with the instruction. The control unit sends it to an address called “X” in the data cache to await the next part of the process.

When the plus (+) key is pressed, the prefetch unit again asks the instruction cache for instructions about what to do with the new data. The prefetch unit translates the code and passes it to the control unit and data cache, which alerts the ALU that an ADD function will be carried out. The process is repeated when the user presses the “2” key.

Next (yes, there’s still more to do), the control unit takes the code and sends the actual ADD command to the ALU. The ALU sums “a” and “b” are added together after they have been sent up from the data cache. The ALU sends the code for “4” to be stored in an address register.

Pressing the equal sign (=) key is the last act the user must execute before getting the answer, but the computer still has a good bit of work ahead of it. The prefetch unit checks the instruction cache for help in dealing with the new keystroke. The resulting instruction is stored, and a copy of the code is sent to the decode unit for processing. There, the instruction is translated into binary code and routed to the control unit. Now that the sum has (finally) been

computed, a print command retrieves the proper address, registers the contents, and displays them. (That involves a separate flurry of activity in the display system, which we won‘t worry about.)

Virtual memory

Virtual memory is the art of using hard disk space to hold data not immediately required by the processor; it is placed in and out of RAM as needed. Although using virtual memory slows the system down as electronic RAM is much faster than a mechanical hard drive.

Real mode versus Protected mode

In simple terms if a processor is said to be running in real mode it will only address the RAM, whereas a processor running in protected mode can access physical (RAM) as well as virtual memory held in a “swap” or “paging” file on the hard disk.

Power supplies

A standard power supply draws power from a local, alternating current (AC) source (usually a wall outlet) and converts it to either 3.3 or 5 volts direct current (DC), for on-board electronics, and 12 volts DC for motors and hard drives. Most PC power supplies also provide the system’s cooling and processor fans that keep the machine from overheating.

There are two main types of power supply for the different types of motherboard in production — AT and ATX.

The older AT-style power supply contained the on/off control and connected to the motherboard through a pair of 6-wire connectors.

The ATX-style motherboard houses the on/off control itself and connects to the power supply through a single 20-pin connector.

Power supplies are rated in terms of the wattage they can supply. It is important to keep in mind that the power supply must produce at least enough energy to operate all the components of the system at one time.

Power supply problems

A role of the PC power supply is to “clean” the power supply as well as convert the voltage. Domestic power supplies often have power surges that can damage the delicate components within a PC. The power supply regulates the power supply so that the PC always a regular feed.

Occasionally a power spike (a surge of very high voltage for a short period of time) will damage (sometimes irreparable) PC components. Businesses therefore often invest in a UPS, or Uninterruptible Power Supply. The USP is connected to the mains and the PC to the UPS. The UPS serves both as a surge protector and also as a backup battery in the event of a power failure.

Expansion bus

As discussed earlier, all devices on a PC are connected to the data bus to enable them to send and receive data to and from the CPU.

Expansion sockets are standardised connectors that allow other devices to be added to a PC that are not already soldered to the motherboard.

Expansion cards that are inserted into an expansion socket connect to the data bus via the expansion bus.

To ease the processing burden on the CPU 3 lot of expansion cards have their own smaller processors dedicated to specific functions. Because these need to run at the same speed as the rest of the PC they often have jumper settings on them which enables the user to set the speed of the card so that it matches the speed of the CPU.

There are several different types of expansion socket in use today.

The first expansion socket was designed by IBM to enable third party manufacturers to produce hardware for the standard IBM PC design. This was called the ISA (Industry Standard Architecture) Bus. It used a 16-bit bus that ran at a top speed of 8.33 MHz.

As CPU power increased the speed of this bus became a hindrance. ISA cards also suffered from being difficult to configure and had a confusing array of jumpers and switches on them.

This lead to the development of the MCA (Micro Channel Architecture) bus.

The MCA used a 32-bit bus and offered a higher speed of 10MHz. All MCA devices also came with a software configuration utility so users did not have to fiddle with hardware switches. However it was totally incompatible with the ISA design and did not prove popular.

IBM responded to the MCA design with a redeveloped version of ISA — EISA (Enhanced Industry Standard Architecture). EISA uses a 32-bit bus and is compatible with older ISA cards.

EISA uses a variation of the ISA slot that accepts older ISA cards, with a two- step design that uses a shallow set of pins to attach to ISA cards and a deeper connection for attaching to EISA cards. In other words, ISA cards slip part-way down into the socket; EISA cards seat farther down.

[image]

Today’s preferred expansion technology is PCI (Peripheral Component Interconnect). It uses a 32-bit bus and operates at around 33 MHz.

PCl uses bus-mastering technology which enables it to regulate the flow of data over the data bus without the CPU’s involvement.

PCI devices are self-configuring which means users do not have to fiddle with hardware switches or jumpers and PCI also allows for IRQ sharing. We will look at IRQs in further details later on. Essentially each device on a PC has its own address which it uses to communicate with the CPU. Normally an IRQ cannot be shared as this leads to loss of information and crashes. PCI enables devices that are not in use at the same time to share lRQs.

However, whilst PCI may seem like the ideal solution, it is important to bear in mind that other manufacturers also use the PCI model, such as Apple. 80 whilst an expansion card may physically fit in your PC, it does not mean it was designed for use with that system.

Accelerated Graphics Port

With the advent of the GUI (Graphical User Interface) and 3D graphics the amount of processing required to display these images soon meant that a new interface had to be developed. The data bus could not handle all of the information being generated by the graphics adapter, so Intel developed the AGP, or Accelerated Graphics Port. This removed all data being generated by the display adapter from the data bus and provided its own pipe directly into the CPU. It also provided a direct path to the system memory for handling graphics. This process is known as DME, or Direct Memory Execute.

Universal Serial Bus (USB)

The latest addition to the PC architecture is the Universal Serial Bus (USB). A USB interface can be added to your PC via an expansion card but most motherboards now have USB ports pre-soldered on them.

USB devices can be connected while the computer is running. Normally the user will be prompted to provide a driver and the device will then be immediately available.

USB can provide either 1.5 Mbps or 12 Mbps data rates depending on the type of device being connected.

I/0 Addresses

As all of the PCs devices are attached to the same data bus, the CPU needs a method of keeping track of them.

Each device is therefore assigned its own address. Normally this is built into the device itself. When the CPU wishes to communicate with a particular device, it sends out the unique code assigned to that device over the data bus, so that device knows that the following data is destined for it. All other devices ignore the data because it is not destined for them.

As mentioned above PCl expansion cards are self—configuring — they determine the best |/O address automatically.

Older ISA cards need to be configured manually with the use of switches and jumpers on the board itself. This requires knowledge of available I/O addresses and can lead to hardware conflicts with other devices.

To view a list of available and used I/O addresses you can use the “winmsd” program (or just “msd” if you running in DOS).

[image]

The l/O address identifies the unique device to the CPU, but to ensure that data is transmitted to and from the CPU reliably, the CPU needs to ensure that no two devices “talk” at the same time.

Controlling the flow of data is called interruption. Every CPU has an interrupt wire. When a device needs to communicate with the CPU it places voltage on this wire. The CPU then stops what it is doing at that time and attends to the device, querying the BIOS to ascertain how it should communicate with the device.

Because the CPU only has one interrupt wire, but it must communicate with several peripheral devices, a secondary chip, known as the Interrupt Controller, or 8259 chip, handles the interrupt requests. If a device needs to interrupt the CPU it goes through the following steps:

1. The device applies voltage to the 8259 chip through its IRQ wire.

2. The 8259 chip informs the CPU, by means of the WT wire, that an interrupt is pending.

3. The CPU uses a wire called an INTA (interrupt acknowledge) to signal the 8259 chip to send a pattern of 1s and 05 on the external data bus. This information conveys to the CPU which device is interrupting.

4. The CPU knows which BIOS to run.

[image]

To see a list of ROS in use on your PC you can run winmsd also:

[image]

Just like |/O addresses, IRQ settings are normally configured automatically by plug and play devices like PCI cards, but older “legacy” devices will need to be configured manually.

The CPU therefore runs the BIOS, the Operating System, any applications that are running as well as the HO addresses and lRQs of the PC’s peripherals. This results in the constant movement of a lot of data. Whilst this consumes a lot of the CPU‘s time, moving data is a relatively uncomplicated task and a waste of the CPU’s resources. Another chip, the DMA chip or 8237 chip, is used to control the data passing from peripherals to the RAM and vice versa. This is known as Direct Memory Access (DMA).

Like l/O addresses, all devices need their own address, or DMA Channel, to be able to communicate with the DMA controller. DMA channel configuration can also be seen in winmsd:

[image]

Like IRQs, devices can inadvertently share the same DMA channel leading to a conflict.

To make the installation of modems and other serial devices easier, all PCs have predefined HO and lRQ values for common ports. Provided a modem Is connected to a port that no other device’ IS using, it will work Here Is a list of predefined values:

Port l/O Address lRQ

COM1 3F8 4

COM2 2F8 3

COM3 5E8 4

COM4 2E8 3

LPT1 378 7

LPT2 278 5

If you buy a network card that by default has an IRQ setting of 4, it will most likely conflict with COM1 — the serial port. If you don’t intend to use the serial port you can disable that port, hence freeing up its l/O address and IRQ. If you want to use both devices then you would need to manually change the setting on the network card itself.

Operating System Basics

Simply put, an operating system is software that controls the operation of a computer. It is the interface between the computer’s hardware and the software applications that run on it. It controls the allocation and usage of hardware resources such as memory, central processing unit (CPU) time, disk space, and peripheral devices.

It‘s important to realize that not all computers have operating systems. The computer that controls the microwave oven in your kitchen, for example, doesn’t need an operating system. It has one set of relatively simple tasks to perform, very simple input and output methods (a keypad and an LCD display), and simple, never—changing hardware to control. For a computer like this, an operating system would be unnecessary baggage, adding complexity where none is required. Instead, the computer in a microwave oven simply runs a single program all the time.

All desktop computers have operating systems. The most common are the Windows family of operating systems (Windows 95, 98, 2000, NT and CE), the UNIX family of operating systems (which includes Linux, BSD UNIX and many other derivatives) and the Macintosh operating systems. There are hundreds of other operating systems available for special-purpose applications, including specializations for mainframes, robotics, manufacturing, real-time control systems and so on.

An OS can be divided into three areas:

The boot files which take control of the computer once the BIOS has initialised

The file system which allows the computer to manage information

The utilities which allow the user to optimize system performance

At the simplest level, an operating system does two things:

It manages the hardware and software resources of the computer system. These resources include things such as the processor, memory, disk space, etc.

It provides a stable, consistent way for applications to deal with the hardware without having to know all the details of the hardware.

[image]

The first task is important as various programs and input methods compete for the attention of the central processing unit (CPU) and demand memory, storage and input/output (l/O) bandwidth for their own purposes. in this capacity, the operating system plays the role of the good parent, making sure that each application gets the necessary resources while playing nicely with all the other applications, and husbanding the limited capacity of the system to the greatest good of all the users and applications.

Within the broad family of operating systems, there are generally four types, categorised based on the types of computers they control and the sort of applications they support. The broad categories are:

Real-Time Operating System (RTOS) – Real-time operating systems are used to control machinery, scientific instruments and industrial systems. An RTOS typically has very little user interface capability, and no end-user utilities, since the system will be a “sealed box” when delivered for use. A very important part of an RTOS is managing the resources of the computer so that a particular operation executes in precisely the same amount of time every time it occurs. In a complex machine, having a part move more quickly just because system resources are available may be just as catastrophic as having it not move at all because the system was busy.

Single-User, Single Task – As the name implies, this operating system is designed to manage the computer so that one user can effectively do one thing at a time. The Palm OS for Palm computers is a good example of a modern single-user, single-task operating system.

Single-User, Multitasking –This is the type of operating system most people use on their desktop and laptop operating systems today. Windows 9x and the MacOS are both examples of an operating system that will let a single user have several programs in operation at the same time. For example, it’s entirely possible for a Windows user to be writing a note in a word processor while downloading a file from the Internet while printing the text of an e-mail message.

Multi-User – A multi-user operating system allows many different users to take advantage of the computer’s resources simultaneously. The operating system must make sure that the requirements of the various users are balanced, and that the programs they are using each have sufficient and separate resources so that a problem with one user doesn’t affect the entire community of users. Unix, VMS and mainframe operating systems, such as MVS, are examples of multi-user operating systems.

Now let’s look at the basic functions of an operating system.

When the power to a computer is turned on, the first program that runs is usually a set of instructions kept in the computer’s Read-Only Memory (ROM) that examines the system hardware to make sure everything is functioning properly. This Power-On Self Test (POST) checks the CPU, memory, and basic input- output systems for errors and stores the result in a special memory location. Once the POST has successfully completed, the software loaded in ROM (sometimes called firmware) will begin to activate the computer‘s disk drives. In most modern computers, when the computer activates the hard disk drive, it finds the first piece of the operating system, the bootstrap loader.

The bootstrap loader is a small program that has a single function: It loads the operating system into memory and allows it to begin operation. in the most basic form, the bootstrap loader sets up the small driver programs that interface with and control the various hardware subsystems of the computer. It sets up the divisions of memory that hold the operating system, user information and applications. It establishes the data structures that will hold the myriad signals, flags and semaphores that are used to communicate within and between the sub-systems and applications of the computer. Finally it turns control of the computer over to the operating system.

The operating system‘s tasks, in the most general sense, fall into six categories:

Processor management

Memory management

Device management

Storage management

Application Interface

User Interface

Let’s look at each of these categories in turn.

Processor Management

The heart of managing the processor comes down to two related issues: First, ensuring that each process and application receives enough of the processor’s time to function properly and, second, using as many processor cycles for real work as is possible. The basic unit of software that the operating system deals with in scheduling the work done by the processor is either a process or a thread, depending on the operating system.

it’s tempting to think of a process as an application, but that gives an incomplete picture of how processes relate to the operating system and hardware. The application you see (word processor or spreadsheet or game) is, indeed, a process, but that application may cause several other processes to begin, for tasks like communications with other devices or other computers. There are also numerous processes that run without giving you direct evidence that they ever exist. A process, then, is software that performs some action and can be controlled – by a user, by other applications or by the operating system.

It is processes, rather than applications, that the operating system controls and schedules for execution by the CPU. in a single-tasking system, the schedule is straightforward. The operating system allows the application to begin running, suspending the execution only long enough to deal with interrupts and user input. interrupts are special signals sent by hardware or software to the CPU. It’s as if some part of the computer suddenly raised its hand to ask for the CPU’s attention in a lively meeting. Sometimes the operating system will schedule the priority of processes so that interrupts are masked, that is, the operating system will ignore the interrupts from some sources so that a particular job can be finished as quickly as possible. There are some interrupts (such as those from error conditions or problems with memory) that are so important that they can’t be ignored. These non-maskable interrupts (NMls) must be dealt with immediately, regardless of the other tasks at hand.

While interrupts add some complication to the execution of processes in a single- tasking system, the job of the operating system becomes much more complicated in a multitasking system. Now, the operating system must arrange the execution of applications so that you believe that there are several things happening at once. This is complicated because the CPU can only do one thing at a time. In order to give the appearance of lots of things happening at the same time, the operating system has to switch between different processes thousands of times a second. Here’s how it happens.

A process occupies a certain amount of RAM. In addition, the process will make use of registers, stacks and queues within the CPU and operating system memory space. When two processes are multi-tasking, the operating system will allow a certain number of CPU execution cycles to one program. After that number of cycles, the operating system will make copies of all the registers, stacks and queues used by the processes, and note the point at which the process paused in its execution. It will then load all the registers, stacks and queues used by the second process and allow it a certain number of CPU cycles. When those are complete, it makes copies of all the registers, stacks and queues used by the second program, and loads the first program.

All of the information needed to keep track of a process when switching is kept in a data package called a process control block. The process control block typically contains an ID number that identifies the process, pointers to the locations in the program and its data where processing last occurred, register contents, states of various flags and switches, pointers to the upper and lower bounds of the memory required for the process, a list of files opened by the process, the priority of the process, and the status of all l/O devices needed by the process. When the status of the process changes, from pending to active, for example, or from suspended to running, the information in the process control block must be used like the data in any other program to direct execution of the task-switching portion of the operating system.

This process swapping happens without direct user interference, and each process will get enough CPU time to accomplish its task in a reasonable amount of time. Trouble can come, through, if the user tries to have too many processes functioning at the same time. The operating system itself requires some CPU cycles to perform the saving and swapping of all the registers, queues and stacks of the application processes. If enough processes are started, and if the operating system hasn‘t been carefully designed, the system can begin to use the vast majority of its available CPU cycles to swap between processes rather than run processes. When this happens, it’s called thrashing, and it usually requires some sort of direct user intervention to stop processes and bring order back to the system.

In a system with two or more CPUs, the operating system must divide the workload among the CPUs, trying to balance the demands of the required processes with the available cycles on the different CPUs. Some operating systems (called asymmetric) will use one CPU for their own needs, dividing application processes among the remaining CPUs. Other operating systems

(called symmetric) will divide themselves among the various CPUs, balancing demand versus CPU availability even when the operating system itself is all that’s running.

Memory management is the next crucial step in making sure that all processes run smoothly.

Memory Management

[image]

When an operating system manages the computer’s memory, there are two broad tasks to be accomplished. First, each process must have enough memory in which to execute, and it can neither run into the memory space of another process, nor be run into by another process. Next, the different types of memory in the system must be used properly, so that each process can run most effectively. The first task requires the operating system to set up memory boundaries for types of software, and for individual applications.

As an example, let’s look at an imaginary system with 1 megabyte of RAM. During the boot process, the operating system of our imaginary computer is designed to go to the top of available memory and then “back up” far enough to meet the needs of the operating system itself. Let‘s say that the operating system needs 300 kilobytes to run. Now, the operating system goes to the bottom of the pool of RAM, and starts building up with the various driver software required to control the hardware subsystems of the computer. In our imaginary computer, the drivers take up 200 kilobytes. Now, after getting the operating system completely loaded, there are 500 kilobytes remaining for application processes.

When applications begin to be loaded into memory, they are loaded in block sizes determined by the operating system. If the block size is 2 kilobytes, then every process that is loaded will be given a chunk of memory that is a multiple of 2 kilobytes in size. Applications will be loaded in these fixed block sizes, with the blocks starting and ending on boundaries established by words of 4 or 8 bytes. These blocks and boundaries help to ensure that applications won’t be loaded on top of one another’s space by a poorly calculated bit or two. With that ensured, the larger question of what to do when the 500 kilobyte application space is filled.

In most computers it’s possible to add memory beyond the original capacity. For example, you might expand RAM from 1 to 2 megabytes. This works fine, but tends to be relatively expensive. It also ignores a fundamental fact of life -— most of the information that an application stores in memory is not being used at any given moment. A processor can only access memory one location at a time, so the vast majority of RAM is unused at any moment. Since disk space is cheap compared to RAM, then moving information in RAM to hard disk intelligently can greatly expand RAM space at no cost. This technique is called Virtual Memory

Management. Operating systems that can only manage the conventional memory space are said to run in real mode. OSes that can manage memory beyond this restriction are said to run in protected mode. Windows 95 and later Microsoft OSes use a technology known as paging which allows for the use of virtual memory. This involves using a portion of the hard disk to write temporary information to. This is considerably slower than conventional RAM, however. The space on the hard disk which is used by the operating system for virtual memory is known as the page file or the swap file.

Device Management

The path between the operating system and virtually all hardware not on the computer‘s motherboard goes through a special program called a driver. Much of a driver’s function is as translator between the electrical signals of the hardware subsystems and the high—level programming languages of the operating system and application programs. Drivers take data that the operating system has defined as a file and translate them into streams of bits placed in specific locations on storage devices, or a series of laser pulses in a printer.

Because there are such wide differences in the hardware controlled through drivers, there are differences in the way that the driver programs function, but most are run when the device is required, and function much the same as any other process. The operating system will frequently assign high priorities blocks to drivers so that the hardware resource can be released and readied for further use as quickly as possible.

One reason that drivers are separate from the operating system is so that new functions can be added to the driver-and thus to the hardware subsystems- without requiring the operating system itself to be modified, recompiled and redistributed. Through the development of new hardware device drivers, development often performed or paid for by the manufacturer of the subsystems rather than the publisher of the operating system, input/output capabilities of the overall system can be greatly enhanced.

Managing input and output is largely a matter of managing queues and buffers, special storage facilities that take a stream of bits from a device, from keyboards to serial communications ports, holding the bits, and releasing them to the CPU at a rate slow enough for the CPU to cope with. This function is especially important when a number of processes are running and taking up processor time. The operating system will instruct a buffer to continue taking input from the device, but to stop sending data to the CPU while the process using the input is suspended. Then, when the process needing input is made active once again, the operating system will command the buffer to send data. This process allows a keyboard or a modem to deal with external users or computers at a high speed even though there are times when the CPU can‘t use input from those sources.

Application Interface

Just as drivers provide a way for applications to make use of hardware subsystems without having to know every detail of the hardware’s operation, Application Program Interfaces (APls) let application programmers use functions of the computer and operating system without having to directly keep track of all the details in the CPU’s operation. Let‘s look at the example of creating a hard disk file for holding data to see why this can be important.

A programmer writing an application to record data from a scientific instrument might want to allow the scientist to specify the name of the file created. The operating system might provide an API function named MakeFile for creating files. When writing the program, the programmer would insert a line that looks like:

MakeFile [1, %Name, 2]

In this example, the instruction tells the operating system to create a file that will allow random access to its data (1), will have a name typed in by the user (%Name), and will be a size that varies depending on how much data is stored in the file (2). Now, let’s look at what the operating system does to turn the instruction into action.

First, the operating system sends a query to the disk drive to get the location of the first available free storage location. With that information, the operating system will create an entry in the file system showing the beginning and ending locations of the file, the name of the file, the file type, whether the file has been archived, which users have permission to look at or modify the file, and the date and time of the file’s creation. Next, the operating system will write information at the beginning of the file that identifies the file, sets up the type of access possible and includes other information that ties the file to the application. In all this information, the queries to the disk drive and addresses of the beginning and ending point of the file will be in formats heavily dependent on the manufacturer and model of the disk drive.

Because the programmer has written his or her program to use the API for disk storage, she doesn’t have to keep up with the instruction codes, data types, and response codes for every possible hard disk and tape drive. The operating system, connected to drivers for the various hardware subsystems, will deal with the changing details of the hardware-the programmer must simply write code for the API and trust the operating system to do the rest.

User Interface

Just as the API provides a consistent way for applications to use the resources of the computer system, a user interface (Ul) brings structure to the interaction between a user and the computer. In the last decade, almost all development in user interfaces has been in the area of the Graphical User Interface (GUI) with two models, Apple’s Macintosh and Microsoft Windows, receiving most of the attention and gaining most of the market share. There are other user interfaces, some graphical and some not, for other operating systems as well.

Unix, for example, has user interfaces called shells that present a user interface more flexible and powerful than the standard operating system text-based interface. Programs such as the Korn Shell and the C Shell are text-based interfaces that add important utilities, but their main purpose is to make it easier for the user to manipulate the functions of the operating system. There are also graphical user interfaces, such as X—Windows and Gnome, that make Unix and Linux more like Windows and Macintosh computers from the user’s point of view.

It‘s important to remember that in all these examples the user interface is a program, or set of programs, that sit as a layer above the operating system itself. The same thing is true, with somewhat different mechanisms, of both Windows and Macintosh operating systems. The core operating system functions, the management of the computer system, lie in the kernel of the operating system. The display manager is separate, though it may be tied tightly to the kernel beneath. The ties between the operating system kernel and the user interface, utilities and other software define many of the differences in operating systems today, and will further define them in the future.

Operating Systems

Now let’s look at some of the common operating systems that have been developed over the years.

MS-DOS (Microsoft Disk Operating System)

Microsoft released MS-DOS in 1981 for lBM PCs and compatibles. It was a single-tasking, single-user operating system with a command line interface.

The IBM PC was a class of personal computer introduced in 1981 which quickly became the de facto standard for PC development. Due to IBM’s decision to release the specification into the public domain a wide range of “compatible” PCs developed by other manufacturers which complied to the same specification soon arose.

DOS has three boot files — “io.sys”, “msdos.sys” and “command.com”

io.sys contains the drivers for the DOS BIOS. It contains drivers for the keyboard and screen, a parallel and serial port, and the system clock.

msdos.sys contains the drivers for the DOS kernel, and provides the functions of all traditional operating systems: file and directory management, character input and output, time and date support, memory management, and country-specific configuration settings.

command.com is the “command interpreter”. it provides what is known as the shell. It is responsible for the handling of internal commands, commands that do not have a “.com” or “.exe” extension (such as copy, for example).

If you enter a command at the prompt, command.com will check whether it is an internal command, if not, then it will search the current folder for a file with a .com or .exe extension matching the command. If it is unable to find such a file, it will return “Bad command or file name”.

Note — DOS also use another file, “config.sys”, which allows you to modify the memory allocation and also specify an alternative to “command.com” for the shell.

These are the system files. Once the system has loaded, DOS then looks for a file called “autoexec.bat”. This file contains user-defined boot settings, the equivalent of the startup folder in Windows.

DOS memory management

A PC’s memory is divided into four distinct sections:

Conventional memory

Upper memory

High memory

Extended memory

The Conventional memory area is limited to 640 KB. DOS can only handle this amount of memory and will load programs into this memory.

As programs became more complicated they became too large to fit into this memory area. Microsoft changed DOS so that it was possible to load DOS itself into the high memory area (HMA), thus freeing up more conventional memory for applications. They also made it possible to load device drivers into the upper memory area (UMA), freeing up even more space.

The DOS file system places an 8-character on the length of filenames and requires that each file has a 3—character extension.

Windows 3.x

Windows 3.x was a multi—tasking Graphical User Interface (GUI) which ran on top of the DOS environment. Windows for Workgroups added limited support for networking.

Windows 9x (95, 98, 988E, ME)

Windows 95 was the first “self-contained” Microsoft operating system that did not use the DOS environment and run on top of it as a shell. lt replaced the DOS file system and included support for filenames up to 255 characters in length and had a radically reworked interface.

The biggest improvement, however, was the introduction of Plug and Play hardware installation. Generally speaking, plug and play refers to the ability of an operating system to automatically configure any new hardware which is added to it. With DOS, users had to manually assign system resources to a device and specify the location of the device’s drivers. Windows 95 was the first Microsoft OS to perform these functions automatically.

Windows 95 also provided support for the FAT16 file system, allowing for partitions of up to ZGB in size.

Windows 98 looks very similar to Windows 95 in terms of appearance, but it provided greater interaction with the Internet, allowing users to access remote files in the same way as they would from their desktop. It also provided support for the FAT32 file system allowing for partitions of up to 2TB in size (Terabytes).

Windows 98SE (Second Edition) provided support for USB devices and also included the Windows Driver Model (WDM). Simply put, the Windows Driver Model contains generic drivers for devices of a similar class, such as modems and printers. This allows manufacturers to write relatively “lightweight” drivers with only the necessary information which is relevant to their hardware.

This allows the manufacturer to write one driver for their device which will work on all OSes that support the WDM. Previously manufacturers had to write separate drivers for Windows 95, 98 and NT.

Windows ME is essentially unchanged from Windows 988E. The interface is slightly improved and there are a number of wizards to simplify tasks such as setting up a home network, but in terms of functionality it is not a radical improvement.

Another dubious improvement introduced with Windows 98, 98 SE and ME was the introduction of ACPI — Advanced Configuration and Power Interface. This has the goal of managing power more efficiently by enabling the OS to automatically power devices on and off as required. But as you will know, this doesn’t always work quite so seamlessly in practice and many problems involving PC cards are resolved by disabling this feature altogether.

Windows NT

Windows NT (New Technology) was first released in 1993. It is a self-contained operating system and NT4 (released in 1996) looks very similar to Windows 95 in terms of the user interface. It is a 32-bit, preemptive multitasking operating system that features networking, symmetric multiprocessing, multithreading, and security.

Windows NT was a radical departure from the thinking behind Windows 95 (hence the name, New Technology) and was designed as a Network Operating System (NOS). It therefore came in two flavours: NT Server and NT Workstation.

The Windows NT architecture is the basis of later Microsoft Operating Systems: Windows 2000, Windows XP and Windows Vista. I will look in more detail at the Windows NT architecture in a later section.

Windows 2000

Windows 2000, like Windows NT before it, is a multi threaded, multitasking 32—bit operating system. Implemented in desktop and several server versions, Windows 2000 focuses overall on improved ease of use, networking, management, reliability, scalability, and security.

it comes in several flavors:

[table]

[table]

Macintosh

A popular series of personal computers introduced by the Apple Computer Corporation in January 1984. The Macintosh was one of the earliest personal computers to incorporate a graphical user interface and the first to use 3.5—inch floppy disks. Despite its user—friendly features, the Macintosh lost market share to PC-compatible computers during the 1990s, but it still enjoys widespread use in desktop publishing and graphics-related applications.

[image]

Santa Cruz Operating Systems (SCO)- SCO Xenix

In 1988 SCO Xenix was released and with it brought the power of UNIX to the lower powered Intel 386 class of PC’s. This multi-user, multi-threaded networked OS was able to work with multi-user enabled software such as Lotus 123 Spreadsheet, WordPerfect Word processor, and FoxBase Relational Database Management System and allow multiple users to share access through smart ASCII based smart terminals to high end peripherals such as high speed dot matrix and laser printers. This worked well in office environments with multiple users needing access to these application programs. This OS had a text based command line interface which made it rather unfriendly to the general user but had more power than dos did at the time. This 08 did have the advantage that it would run the most popular office applications at that time right out of the box.

Introduction to SMS Messaging

This guide will look at the evolution of mobile messaging from its early inception as SMS up to the current technological advances of MMS.

It will also explain the mechanics behind message delivery and how networks are using the potentials of messaging to increase their revenues.

SMS

SMS (Short Message Service, or “texting” as it is more commonly known) has long been a feature of GSM, and has been available since 1992.

It is a common form of communication and today accounts for 15% of mobile operators’ revenue.

[image]

The earliest SMS service enabled users to send messages only to other users within the same network.

The ability to send SMS messages between networks — nationally and internationally — was not possible until the late 1990s.

Today almost 15 billion messages are delivered globally each month.

A single SMS message can contain up to 160 characters. Some mobile devices allow several messages to be “concatenated” (joined together) allowing a maximum of 9 messages to be sent, representing a total message length of 1358 characters.

These messages will be delivered as a constant stream of text and the recipient will not notice any breaks in the text provided that their handset also supports concatenation. If it does not then the user will see 9 messages with random characters at the beginning and end of each message.

Note — the sender is still charged for sending 9 messages.

SMS messages can be received while a phone call is in progress, as all SMS traffic is sent via the GSM signalling channel. This enables the network to make a more efficient use of scarce radio resources.

The signalling channel, known as 887, enables the mobile device to communicate with the base station for the purposes of location management and call setup.

[image]

Smart Messaging

In 1999, Nokia announced a development of the conventional SMS standard which allowed for additional data to be sent with a message, including:

* Ringtones

* Business cards

* Calendar appointments

* WAP or Internet settings

* Picture messages (comprised of three messages)

This heralded the first development of multimedia messaging, but as it was a proprietary Nokia standard messages could only be exchanged between those Nokia handsets which supported the standard.

It did mean, however, that handsets could be provisioned for new services such as WAP automatically without user intervention. The sending of service settings via SMS is called OTA provisioning (Over The Air - a service which WDS now provides with Mouse2Mobile)

Enhanced Messaging Service (EMS)

Launched in 2000, EMS was Ericsson’s answer to Nokia’s Smart Messaging platform. It did not support OTA provisioning nor calendar or contact information transfer.

But it did provide support for pictures, sounds and stylised text. This meant that messages could be personalised with emoticons — small graphics to express a sentiment, and users could personalise their phones with logos, ringtones and screensavers.

How are SMS messages sent?

Standard SMS, Smart Messages and EMS messages are all sent in the same way. All messages pass through a component of the network called the SMSC (Short Message Service Centre)

The SMSC consults with the HLR and VLR to ascertain the current location of the user, and sends the information to the MSC for routing across the network to the target device.

[image]

The SMSC is responsible for storing messages if delivery is not immediately possible (if the target device is switched off or is out of range).

Delivery attempts are repeated at a pre—defined interval. Some devices allow a “message validity” to be set, so that a message will expire if not delivered within a certain time frame.

Both Smart Messaging and EMS brought new life to the SMS service with sounds and graphics, and added impetus to further development of the service, which has now culminated in the development of MMS — the Multimedia Messaging Service.

[image]

MMS will enable multiple forms of media to be combined in a single message —text, graphics, animation and audio samples.

Later stages of the service will develop with the evolution of GSM as a whole and allow for the inclusion of motion video.

This could be delivered in two ways — 30 seconds of motion video could recorded with the terminal device’s built-in camera and included in the body of a message.

Alternatively, the message could contain a link to a web server which would live stream video to the terminal.

How are MMS messages delivered?

As you will remember SMS messages are delivered over the GSM network signaling channel. Due to the larger amount of data transfer involved in sending an MMS message, this is no longer possible.

Both the signaling channel and conventional voice and data channels will be required.

[image]

Whereas SMS messages are delivered using the 887 protocol, MMS messages will use a new protocol — Wireless Session Protocol, or WSP.

WSP is already used by WAP services for the routing of data. It handles the sending of a WAP page request to the WAP server that holds the page, and sending the page back to the terminal device.

This will guarantee compatibility with a wide range of existing handsets. However, although MMS uses protocols in use by WAP, MMS does not use the browser function of the handset.

If the mobile network operator implements an MMS email gateway, it is possible to send MMS messages from the mobile device to an Internet email recipient.

All MMS messages are encoded as a MIME (Multipurpose Internet Mail Extension) message and sent over the network using SMTP. This guarantees that the message will be viewable in almost any email program.

[image]

Due to the amount of data involved in the sending of an MMS message, it is not suitable for sending over conventional CSD. HSCSD or GPRS carriers are preferred.

MMS Network Components

In order to provide an MMS service, the network operator needs to add several components to the existing GSM network infrastructure. Together these are called the Multimedia Message Service Environment.

[image]

MMSC — Multimedia Messaging Service Centre

As with conventional SMS, the MMSC is responsible for storing the message until it has been successful received by the intended recipient. It comprises two pieces of hardware — the MMS Proxy Relay and the MMS Server. The Server stores the message, the Proxy Relay forwards messages or sends message notifications to terminal devices.

Application Gateway

The Application Gateway allows Internet—based service providers to send content to mobile devices in the form of an MMS message. The gateway can perform the necessary format conversion where necessary (from HTML to an MMS-compatible format, for example), and can resize images for display on smaller screens.

User Database

The User Database holds information on the capabilities of a user’s mobile device and what services the user has subscribed to. User’s details are stored in the database in a User Agent Profile (UAP)

The user agent profile might contain such information as:

Maximum size of message supported in bytes

Maximum size of image supported in bytes

Supported content formats

Supported character sets

Accepted languages

Accepted encoding formats

Screen size

Processor type

Users may be able to access their profile via a web-based interface and specify custom user settings, such as setting up filters to automatically discard undesired messages, specifying which message formats they wish to have forwarded to their device, and which message types they wish saved for later retrieval.

Message Store

Also known as the MMS Terminal Gateway. This provides MMS support for devices that do not themselves support MMS messaging. This will store the MMS message and send notification via conventional SMS to the device to indicate that a message is waiting for the user. The user will then need to connect to the store via an Internet connection, from their PC for example.

Such devices are known as legacy devices.

The Message Store could also serve as a personal album for users. Each user will be assigned a certain amount of storage space and can opt to save favourite messages here for later retrieval.

Multimedia Email Gateway

The MMS Email Gateway provides the necessary functionality for MMS messages to be sent to email recipients. Similarly, incoming emails can be converted to MMS format for delivery to mobile terminals.

Multimedia Voice Gateway

The Voice Gateway enables the user to record voice samples and send them directly to the recipient via the MMS service. Serving a similar function to voicemail, this is a more personalised and much faster service.

WAP Gateway

The WAP Gateway enables multimedia content to be compressed prior to transmission over the wireless network to increase network efficiency. Most network operators will already provide WAP gateways for their existing WAP services. In this context it is responsible for converting HTML pages held in the “wired” Internet, into WML format for interpretation by the “wireless” browser on the user’s handset.

MMS — from composition to delivery

In its initial phase, an MMS message will resemble a miniature slide slow. A slide could contain any of the following:

Text in various sizes and colours

Animated colour graphics

Digital photographs

Sampled digital audio

Synthesised audio (beeps or tones)

MMS messages M” be created using a programming language optimised for the new format, known as SMIL (Synchronised Multimedia Integration Language).

Pronounced “smile”, the language is similar in format to HTML and WML and contains information governing the content and timing of the presentation.

[image]

Each MMS message will contain one slide. Each slide may be comprised of a graphic region and a text region. Timing options can be set to that one or the other appears to the recipient first.

[image]

The orientation of the message can be specified also — landscape or portrait — to best harness the screen resolution of the recipient’s device.

In Phase 1, MMS messages will be delivered using existing WAP protocols and network infrastructure. Due to limitations of this technology MMS messages will initially be limited to a maximum of 30 Kb.

The message itself will be “built” as follows:

[image]

Header information

SMIL code

The image for the first slide

The image for the second slide

The text for the first slide

The text for the second slide

The audio for the first slide

The audio for the second slide

The SMIL code will contain the slide layout and also information governing the timing of presentation of the different message elements.

Once complete, all of the elements are compressed and saved as a single MIME “object”, guaranteeing efficient data transfer over the network as well as compatibility with a wide range of both other mobile devices and also other devices connected to the wired Internet.

[image]

To send the message the sender will need to enter the recipient’s voice number or email address.

The MMS message is then sent from the sender’s mobile device to the network’s MMSC using the WSP protocol.

The message is then stored at the MMSC and a notification message is then sent to the recipient’s terminal.

The message is sent using “PUSH” technology. This was developed to enable WAP content providers to send unsolicited messages to users without the user’s specific request.

To avoid receiving junk messages users can now specify from which gateways and servers they wish to receive messages — unless the user has granted access to the correct gateway then messages may be blocked.

Provided the recipient’s terminal can receive an MMS message, it will send a request message to the MMSC to download the message.

The MMSC then sends the message to the terminal, and once it receives confirmation of the message’s receipt, it is removed from the message store.

It is possible for the user to specify that a conventional SMS be sent to the terminal to indicate that an MMS is waiting for them, so that the user can then decide whether or not to download the message.

Due to the “always on” nature of GPRS connections, it is the preferred “bearer” service for MMS messaging, enabling seamless message delivery with little or no user intervention.

Introduction to TCP/IP

You will all (hopefully!) have heard of TCP/IP and know that it is somehow used by the Internet. But what is it really and how does it work? This chapter will give an introduction to how computers exchange information on a network and how networks can be connected together to form an internetwork. I will also explain what TCP and IP are and how they work together. This chapter assumes no prior knowledge on any of these subjects, but is intended as a brief introduction only.

Sections

An overview of TCP

An introduction to networking

The OSI Reference Model

Bridges, Routers and Gateways

Network Addressing

An overview of IP

IP Addressing

The IP header

DHCP

SLIP and PPP

A Datagram’s life

What is NAT and how does it work?

What are ports and sockets?

Appendices

An overview of TCP

TCP stands for Transmission Control Protocol.

TCP is actually a suite of protocols. It was developed during the 1970s and 1980s by the US Department of Defence and is designed to allow dissimilar systems (such as Windows and Unix) to communicate with each other. It is an open system which means it is completely hardware and software independent.

Its main function is to translate the data created by applications requiring access to the network (be it email, FTP or whatever) into a common form. It is also responsible for monitoring the transmission of data and correcting any errors that occur.

Together with IP (Internet Protocol) it can ensure that data is routed across the network over the shortest route possible. It is also able to automatically reroute data if one or more nodes on the network become damaged.

TCP/IP was first developed in 1969 by the US Department of Defence. Prior to 1969 the entire intelligence and defence reflex computer system was held in one location. This was considered to be a potentially vulnerable target, so millions of dollars was invested in creating what became known as DARPAnet (Defence Advanced Research Projects Agency Network). DARPAnet became one of the largest Wide Area Networks in the world, spanning several US states and enabling data to be replicated between sites.

In order to understand how TCP operates and how it interacts with other protocols it is necessary to understand how networks communicate.

An introduction to networking

In simple terms, a network is a set of computers and peripherals (printers, scanners etc) that are connected by some medium. The connection can be direct (via cable, IR or radio), or indirect (via a modem).

Whatever the connection method, all networks communicate following a set of rules (protocols).

Networks can be confined to a room, a building or can be scattered across many different continents. The geographical layout of a network is known as the network topology.

A group of connected computers in a building is known as a Local Area Network (LAN). Two different LANs in different cities could be connected to form a Wide Area Network (WAN). Although these networks are miles apart, as far as the Internet is concerned it is one entity as it has a single ”domain”. Taking WDS as an example, the company has a domain of wdsglobal.com, but has several subdomains of uk, us, au and za. Each region has its own local area network, but each local area network can communicate with each other, forming a wide area network.

[image]

Computers can communicate over networks in many ways and for many reasons, but almost all of the process to transfer information from one machine to another is unconcerned with the nature of the data itself. By the time the data generated by the transmitting computer reaches the cable or other medium, it has been reduced to signals that are native to that medium. These might be electrical voltages, for a copper cable network; pulses of light, for fiber optic; or infrared or radio waves. The receiving computer translates these signals back into binary data and it is interpreted by the relevant application, be it an email client, web browser or whatever.

Baseband and broadband

In most cases, LANs use a shared network medium. The cable connecting the computers can carry one signal at a time, and all of the systems take turns using it. This type of network is called a baseband network.

Baseband networks transmit data in small chunks of data called packets. For this reason they are known as packet-switched.

The alternative to packet-switching is circuit-switching. Circuit-switching means that the two systems wanting to communicate establish a circuit before they transmit any information. That circuit remains open throughout the duration of the exchange, and is only broken when the two systems are finished communicating. This is an impractical solution for computers on a baseband network, because two systems could conceivably monopolise the network medium for long periods of time, preventing other systems from communicating. Circuit switching is more common in environments like the public switched telephone network (PSTN), in which the connection between your telephone and that of the person you’re calling remains open for the entire duration of the call.

A broadband network is the opposite of a baseband network — it can carry multiple signals on the same medium simultaneously.

Half duplex and full duplex

When two computers communicate over a LAN, data typically travels in only one direction at a time, because the baseband network used for most LANs supports only a single signal. This is called half-duplex communication. By contrast, two systems that can communicate in both directions simultaneously are operating in full-duplex mode. An example of half duplex communication would be a CB radio, where each party has to wait for the other to finish before transmitting. An example of full duplex would be a telephone conversation.

[image]

Clients and servers

A sewer is simply a computer that provides services to other computers, for example a file server or a print server. A client is the computer that accesses the server to make use of these services.

All computers can perform both roles. On most networks servers are more powerful machines with a single role to perform. Networks comprised of machines which perform both client and server functions are called peer-to-peer networks.

The OSI Reference Model

In 1983, the International Organization for Standardization (ISO) and what is now the Telecommunications Standardisation Sector of the International Telecommunications Union (lTU-T) published a document called “The Basic Reference Model for Open Systems Interconnection.”

The OSI model defines the networking process and breaks it down into 7 layers. This is a theoretical construction designed to make the process easier to understand. At the top of the model is the application that requires access to a resource on the network, and at the bottom is the network medium itself. As data moves down through the layers of the model, the various protocols operating there prepare and package it for transmission over the network. Once the data arrives at its destination, it moves up through the layers on the receiving system, where the same protocols perform the same process in reverse.

The seven layers are shown below:

[image]

The Application, Presentation and Session layers are known as the upper layers and are responsible for the user interface.

They are independent of (and oblivious to) the four layers below them.

The lower layers deal with the transmission of data.

They don’t differentiate between applications, they are concerned with data and where to send it.

The protocols operating at each layer of the OSI model add a header (and in one case a footer) to the data received from the layer above it. For example, when an application generates a request for a network resource, it passes the request down through the protocol stack. When the request reaches the transport layer, the transport layer protocol adds its own header to the request. The header consists of fields containing information that is specific to the functions of that protocol.

The transport layer protocol, after adding its header, passes the request down to the network layer. The network layer protocol then adds its own header in front of the transport layer protocol’s header. The original request and the transport layer protocol header thus become the payload for the network layer protocol. This entire construct then becomes the payload for the data-link layer protocol. which typically adds both a header and a footer. The final product, called a packet, is then ready for transmission over the network. After the packet reaches its destination, the entire process is repeated in reverse. The protocol at each successive layer of the stack (traveling upwards this time) removes the header applied by its equivalent protocol in the transmitting system. When the process is complete, the original request arrives at the application for which it was destined in the same condition as when it was generated.

The process by which the protocols add their headers and footer to the request generated by the application is called data encapsulation.

[image]

The procedure is functionally similar to the process of preparing a letter for mailing. The application request is the letter itself, and the protocol headers represent the process of putting the letter into an envelope, addressing, stamping, and mailing it.

Let’s look at each of the layers in a little more detail.

The Physical Layer

At the bottom of the OSI model (or “protocol stack”) is the physical layer. This is the hardware of the computer’s network interface adapter, be it an Ethernet card, an infrared port or whatever. The physical layer is often determined by the protocol used in the data link layer. For example Ethernet is a data link layer protocol but supports many different types of cabling.

The Data Link Layer

The protocol operating on the data link layer is the conduit between the computer’s networking hardware and its networking software (in the network layer). The protocols on the network layer pass data down to the data link layer which packages it for transmission over the network. At the receiving machine the data link layer protocol receives data from the physical layer and passes it up to the network layer after removing the header and footer generated by the data link layer protocol on the other machine.

[image]

[image]

It is important to understand that data-link layer protocols are limited to communications with computers on the same LAN. The hardware address in the header always refers to a computer on the same local network, even if the data’s ultimate destination is a system on another network.

The other primary function of the data link layer is to determine what protocol on the network layer created the frame — a computer can have several different protocols operating on the network layer. It is up to the data link layer to ensure that frames created by one protocol get delivered to the same protocol on the receiving machine.

The data link layer is also responsible for error detection. This is accomplished by the Cyclic Redundancy Check (CRC) which is included in the footer. The CRC is a means of checking that the packet has been received successfully. It takes the sum of all the 15 in the payload and adds them together. The result is stored as a hexadecimal value in the trailer. The receiving device adds up the is in the payload and compares the result to the value stored in the trailer. If the values match, the packet is good. But if the values do not match, the receiving device sends a request to the originating device to resend the packet.

[image]

[image]

The Network Layer

Data link layer protocols only function on the local area network. Network layer protocols are what is known as “end-to-end” protocols. They are responsible for communicating with the destination machine to ensure that transmitted data is received successfully.

The most commonly used network layer protocol is IP, Internet Protocol. As a frame moves across the Internet, the data link layer protocol may change several times (as the physical medium of the network it is on changes), but the network layer protocol will remain intact throughout the trip.

Like the data link layer protocol, the network layer protocol applies a header to the data it receives from the layer above.

[image]

The unit of data created by the network layer is called a datagram.

[image]

[image]

As mentioned earlier, network layer protocols are responsible for ensuring the datagram‘s entire journey to the destination machine. This process is known as routing.

The individual networks that make up the Internet are connected by Routers. A router’s function is to examine each packet that passes through it and determine whether or not it is destined for a machine on the network it is connected to or, if not, forward it to another router. To do this the router must examine the IP address of the packet, which means passing the datagram up to the network layer:

[image]

The router then processes the packet and sends it back down to the data link layer to be transmitted to its next destination.

[image]

The Transport Layer

Transport layer protocols work very closely with network layer protocols. IP is a network layer protocol. The most common transport layer in use today is TCP (Transmission Control Protocol).

There are two types of protocol in the transport layer — connection-oriented and connection/ass.

A connection-oriented protocol is one in which the two communicating systems exchange messages to establish a connection before they transmit any application data. This ensures that the systems are both active and ready to exchange messages. TCP, for example, is a connection’ oriented protocol. When you use a Web browser to connect to an Internet sewer, the browser and the server first perform what is known as a three-way handshake to establish the connection. Only then does the browser transmit the address of the desired Web page to the server. When the data transmission is completed, the systems perform a similar handshake to break down the connection.

Connection-oriented protocols also provide additional services such as packet acknowledgment, data segmentation, flow control, and end-to-end error detection and correction. Because of these services, connection-oriented protocols are often said to be reliable. The drawback of this type of protocol is that it greatly increases the amount of control data exchanged by the two systems, effectively slowing the connection.

Connectionless protocols (such as UDP, User Datagram Protocol), do not offer these services and are therefore said to be unreliable. But as they transmit much less data they are often used for network signalling and control messages.

The Session Layer

Up until now all the protocols discussed have been involved in the communication of data. The Session layer is not concerned with data transmission, rather it coordinates the exchange of data between the lower layers and the upper layers.

The Presentation Layer

The presentation layer is responsible for determining what application the “raw” data from the lower layers is destined for, and delivering it to the corresponding port number (we’ll look at ports in more detail later).

The Application Layer

The application layer is the entrance point that programs use to access the OSI model and utilise network resources. Most application layer protocols provide services that programs use to access the network, such as the Simple Mail Transfer Protocol (SMTP), which most e-mail programs use to send e-mail messages. In some cases, such as the File Transfer Protocol (FTP), the application layer protocol is a program in itself.

Remember that the OSI seven-layer model is a theoretical construct, a guide to understanding the procedure for transmitting data over a network. In the TCP model application layer protocols include the session and presentation layer functions. As a result, the TCP protocol stack consists of four layers - application, transport. network and data-link.

Bridges, Routers and Gateways

A bridge is a piece of hardware normally with two network interface adapters used to connect to network segments together. The bridge analyses the destination address of all packets and determines which of the two networks the packet is destined for. A bridge can only be used where the two protocols use the same data link layer protocol (ie the same network technology).

A router is simply a machine on a network (a node) that forwards packets around the network.

A gateway is a device that performs routing functions, but can perform protocol translation from one network to another if necessary.

Network Addressing

All networks use a form of addressing to transfer data from one machine to another. Network addresses are analogous to mailing addresses in that they tell a system where to deliver data. Three terms commonly used in networking relate to addressing: name, address, and route.

A name is a specific identification of a machine, a user, or an application. It is usually unique and provides an absolute target for the data. An address typically identifies where the target is located, usually its physical or logical location in a network. A route tells the system how to get data to the address.

Each device on a network that communicates with others has a unique physical address, sometimes called the hardware address. The hardware address is also known as the MAC address (Media Access Control) and is burned into the network adapter card itself. Legally no two adapter cards can have the same MAC address; therefore no two machines on the same network can have the same address. The MAC address consists of a 3-byte value called an organizationally unique identifier (OUI), which is assigned to the adapter’s manufacturer by the IEEE, plus another 3-byte value assigned by the manufacturer itself.

Ethernet, like all data-link layer protocols, is concerned only with transmitting packets to another system on the local network. If the packet’s final destination is another system on the LAN, the Destination Address field contains the address of that system’s network adapter. If the packet is destined for a system on another network, the Destination Address field contains the address of a bridge or gateway on the local network that provides access to the destination network. It is then up to the network layer protocol to supply a different kind of address (such as an Internet Protocol [IP] address) for the system that is the packet’s ultimate destination. A typical Ethernet packet looks like this:

[image]

This is the addressing system used on local networks. The addressing system used on an Ethernet network cannot be used to send a packet to a machine on another network. Ethernet is a data link layer protocol. Only network layer protocols can address packets to other networks (as they are end-to-end protocols, remember?)

As mentioned earlier lP (Internet Protocol) is the most common network layer protocol in use.

Internet Protocol

Despite what I have just said above, although the word “Internet” appears in this protocol’s name, it is not limited to use on the Internet, it can be used on dedicated networks that have no connection to the Internet as well.

IP is responsible for sending the data provided by the transport layer protocol (TCP most commonly) to its destination. It can work with any transport layer protocol, it does not have to be TCP.

lP was developed by a man called Vint Cerf who was part of the DARPAnet project.

The destination address in the header created by IP is the address of the target machine. Along the way to this machine the datagram may be encapsulated with headers from several other protocols, but the lP datagram itself will never be changed en route. The process is similar to the delivery of a letter by the post office, with IP functioning as the envelope. The letter might be placed into different mailbags and transported by various trucks and planes during the course of its journey, but the envelope remains sealed. Only the addressee is permitted to open it and make use of the contents.

Now let’s get a bit technical!

The IP Header

When the IP protocol receives data from the transport layer it adds a 20-byte header to it:

[image]

The header provides the following information:

• Version (4 bits) — the version of IP that created the datagram (the version in current use is 4, but version 6 is being developed).

• IHL (Internet Header Length) (4 bits) — tells the receiving system the length of the IP header.

• Type of Service (1 byte) — specifies the priority for the datagram. Rarely used this field can be used to specify the importance of the datagram so that a router will forward this as soon as it receives it, rather than buffering it.

• Total Length (2 bytes) — tells the receiving system the length of the entire datagram (in case it will need to fragment it).

• ldentification (2 bytes) — this field numbers the datagram so that the receiving machine can ensure datagrams are reassembled in the correct order.

• Flags (3 bits) — used to regulate the fragmentation process.

• Fragment offset (13 bits) — if a datagram is fragmented, this field identifies the fragments place in the datagram for reassembly.

• Time To Live (1 byte) - This field specifies the number of networks that the datagram should be permitted to travel through on the way to its destination. Each router that forwards the datagram reduces the value of this field by one. If the value reaches zero, the datagram is discarded. This is important otherwise datagrams could in theory roam the Internet forever.

• Protocol (1 byte) — This field contains a code that identifies the protocol that generated the information found in the Data field.

• Header Checksum (2 bytes) — This field contains a Checksum value computed on the IP header fields only (and not the contents of the Data field), for the purpose of error detection.

• Source IP Address (4 bytes) — This field specifies the IP address of the system that generated the datagram.

• Destination IP Address (4 bytes) — This field specifies the IP address of the system for which the datagram is destined.

• Options (variable) — This field is present only when the datagram contains one or more of the 16 available lP options. The size and content of the field depends on the number and the nature of the options. Options include “timestamping” the datagram, allowing the IP header to increase to include the IP address of all routers it has passed through among others

• Data (variable) — This field contains the information generated by the protocol specified in the Protocol field. The size of the field depends on the data-link layer protocol used by the network over which the system will transmit the datagram.

IP Addressing

The IP protocol is unique in that it has its own unique self-contained addressing system that it uses to identify computers on networks of almost any size.

IP uses a 32-bit addressing scheme which contains both a network ID and a host ID, to identify both the network which the target machine is on as well as the target machine’s logical location on that network.

[image]

The address is notated using four decimal numbers ranging from O to 255, separated by periods, as in 192.168.1.44. This is known as dotted decimal notation. This provides a total range of some 4,294,967,296 unique addresses.

Unlike hardware addresses, which are hardcoded into network interface adapters at the factory, network administrators must assign IP addresses to the systems on their networks. It is essential for each network interface adapter to have its own unique IP address; when two systems have the same IP address, they cannot communicate with the network properly.

As mentioned earlier, IP addresses consist of two parts: a network identifier and a host identifier. All of the network interface adapters on a particular subnet have the same network identifier, but a different host identifier. For systems that are on the Internet, the network identifiers are assigned by a body called the Internet Assigned Numbers Authority (IANA). This is to ensure that there is no address duplication on the Internet. When an organization registers its network, it is assigned a network identifier, and it is then up to the network administrators to assign unique host identifiers to each of the systems on that network. This two-tier system of administration is one of the basic principles of the Internet.

IP Address Classes

The most complicated aspect of an IP address is that the division between the network identifier and the host identifier is not always in the same place. A hardware address, for example, consists of 3 bytes assigned to the manufacturer of the network adapter and 3 bytes which the manufacturer itself assigns to each card. IP addresses can have various numbers of bits assigned to the network identifier, depending on the size of the network.

Class A addresses are for large networks that have many machines. The 24 bits for the local address (also frequently called the host address) are needed in these cases. The network address is kept to 7 bits, which limits the number of networks that can be identified.

Class B addresses are for intermediate networks, with 16-bit local or host addresses and 14-bit network addresses.

Class C networks have only 8 bits for the local or host address, limiting the number of devices to 256. There are 21 bits for the network address.

Finally, Class D networks are used for multicasting purposes, when a general broadcast to more than one device is required. The lengths of each section of the IP address have been carefully chosen to provide maximum flexibility in assigning both network and local addresses.

[image]

There is another class of address, Class E, that is as yet unused.

DHCP — Dynamic Host Control Protocol

Because each machine on a network must have a unique IP address to ensure that data is sent to the correct destination, network administrators used to have an arduous task of maintaining large networks. Microsoft developed a protocol to address this specific issue and to make IP assignment an automatic process. DHCP is their solution.

DHCP consists of three components: a client, a server, and the protocol that they use to communicate with each other. Most TCP/IP implementations these days have DHCP integrated into the networking client, even if the operating system doesn’t specifically refer to it as such. On a Microsoft Windows 98 system, for example, when you select the “Obtain An IP Address Automatically” radio button in the TCP/IP Properties dialog box, you are actually activating the DHCP client. The DHCP server is an application that runs on a computer and exists to service requests from DHCP clients.

The core function of DHCP is to assign IP addresses. This is the most complicated part of the service, because an IP address must be unique for each client computer. The DHCP standard defines three types of IP address allocation, which are as follows:

Manual allocation An administrator assigns a specific IP address to a computer in the DHCP server and the server provides that address to the computer when it is requested.

Automatic allocation The DHCP server supplies clients with IP addresses taken from a common pool of addresses, and the clients retain the assigned addresses permanently.

Dynamic allocation The DHCP server supplies IP addresses to clients from a pool on a leased basis. The client must periodically renew the lease or the address returns to the pool for reallocation.

SLIP and PPP

SLIP (Serial Line Internet Protocol) and PPP (Point to Point Protocol) are data link layer protocols. They are part of the TCP/IP protocol suite and are not designed to connect machines on a LAN. Despite being data link layer protocols they are in fact end-to-end protocols. They are designed to connect one system to another using a dedicated connection, such as a telephone line. Because the medium is not shared there is no need for a MAC mechanism, and because there are only two systems involved there is no need to address the packets.

SLIP is so simple it hardly deserves to be called a protocol. It is designed to transmit signals over a serial connection (which in most cases means a modem and a telephone line) and has very low control overhead, meaning that it doesn’t add much information to the network layer data that it is transmitting. Compared to the 18 bytes that Ethernet adds to every packet, for example, SLIP adds only 1 byte. Of course, with only 1 byte, SLIP can’t provide functions like error detection, network layer protocol identification, security, or anything else. Because of its limitations it is rarely used today, having largely been replaced by PPP.

PPP is, in most cases, the protocol you use when you access the Internet by establishing a dial-up connection to an Internet Service Provider (ISP). PPP is more complex than SLIP and is designed to provide a number of services that SLIP lacks. These include the ability of the systems to exchange IP addresses, carry data generated by multiple network layer protocols (which is called multiplexing), and support different authentication protocols. Still, PPP does all this using only a 5-byte header, larger than SLIP but still less than half the size of the Ethernet frame.

Establishing a PPP connection

In order to reduce the amount of data transmitted the PPP frame does not contain all of the information necessary to fulfil all of the functions described above. Instead the PPP protocol involves a lengthy handshaking procedure at the beginning of the connection. Because the packet is smaller the connection is more efficient. Whilst it is necessary in Ethernet for every packet to contain the address of the target machine, in a PPP connection this is not necessary as there are only two machines involved.

The connection is established as follows:

1. At the beginning the link is dead until one of the two systems runs a program, such as one that causes a modem to dial.

2. Link Establishment —Once the physical layer connection is established, one system generates a PPP frame containing a Link Control Protocol (LCP) Request message. The systems use the LCP to negotiate the parameters they will use during the rest of the PPP session. The message contains a list of options, such as the use of a specific authentication protocol, header compression, network layer protocols, and so on. The receiving system can then acknowledge the use of these options or deny them and propose a list of its own. Eventually, the two systems agree on a list of options they have in common.

3. Authentication —If the two systems have agreed to the use of a particular authentication protocol during the link establishment phase, they then exchange PPP frames containing messages particular to that protocol in the Data field. Systems commonly use the Password Authentication Protocol (PAP) or the Challenge Handshake Authentication Protocol (CHAP), but there are other authentication protocols as well.

4. Link Quality Monitoring —If the two systems have negotiated the use of a link quality monitoring protocol during the link establishment phase, the exchange of messages for that protocol occurs here.

5. Network Layer Protocol Configuration —For each of the network layer protocols that the systems have agreed to use, a separate exchange of Network Control Protocol (NCP) messages occurs at this point.

6. Link Open —Once the NCP negotiations are complete, the PPP connection is fully established, and the exchange of packets containing network layer application data can commence.

7. Link Termination— When the two systems have finished communicating, they sever the PPP connection by exchanging LCP termination messages, after which the systems return to the Link Dead state.

Summary — A datagram’s life

You have now looked at how TCP constructs a datagram, and how IP is used to route it to the destination machine — be it on an Ethernet network, or over the Internet. Let’s examine how a datagram passes along the Internet in a bit more detail.

When an application must send a datagram out on the network, it performs a few simple steps. First, it constructs the IP datagram within the legal lengths stipulated by the local IP implementation. The checksum is calculated for the data, and then the IP header is constructed. Next, the first hop (machine) of the route to the destination must be determined to route the datagram to the destination machine directly over the local network, or to a gateway if the internetwork is used. If routing is important, this information is added to the header using an option. Finally, the datagram is passed to the network for its manipulation of the datagram.

As a datagram passes along the internetwork, each gateway stops the packet and removes the outer header. The IP protocol on the network layer on the gateway’s network interface calculates the checksum and verifies the integrity of the datagram. If the checksums don’t match, the datagram is discarded and an error message is returned to the sending device. Provided it does, next the TTL field is decremented and checked. If the datagram has expired, it is discarded and an error message is sent back to the sending machine. After determining the next hop of the route, either by analysis of the target address or from a specified routing instruction within the Options field of the IP header, the datagram is rebuilt with the new TTL value and new checksum.

If fragmentation is necessary because of an increase in the datagram length or a limitation in the software, the datagram is divided, and new datagrams with the correct header information are assembled. If a routing or timestamp is required, it is added as well. Finally, the datagram is passed back to the network layer.

When the datagram is finally received at the destination device, the system performs a checksum calculation and—assuming the two sums match—checks to see if there are other fragments. If more datagrams are required to reassemble the entire message, the system waits, meanwhile running a timer to ensure that the datagrams arrive within a reasonable time. If all the parts of the larger message have arrived but the device can’t reassemble them before the timer reaches 0, the datagram is discarded and an error message is returned to the sender. Finally, the IP header is stripped off, the original message is reconstructed if it was fragmented, and the message is passed up the layers to the upper layer application. If a reply was required, it is then generated and sent back to the sending device.

It’s a miracle it works at all!

What is NAT and how does it work?

NAT stands for Network Address Translation.

If I were to publish this article on my website and you were to view from your work machine then you would be using NAT right now!

When the Internet was first “developed”, everyone thought that there were plenty of addresses to cover any need. Theoretically, you could have 4,294,967,296 unique addresses (2”). The actual number of available addresses is smaller (somewhere between 3.2 and 3.3 billion) because of the way that the addresses are separated into classes, and because some addresses are set aside for multicasting, testing or other special uses.

With the explosion of the Internet and the increase in business networks, the number of available IP addresses is simply not enough. The obvious solution is to redesign the address format to allow for more possible addresses. This is being developed (called IPv6), but will take several years to implement because it requires modification of the entire infrastructure of the Internet.

Network Address Translation allows a single machine to act as a gateway between a private network and the public Internet. Therefore only one IP address needs to be visible to the Internet and can represent a whole network of other machines.

A Local Area Network uses IP addresses internally. Most of the network traffic in a LAN is local, so it doesn’t travel outside the internal network. A LAN can include both registered (global) and unregistered (local) IP addresses. Of course, any computers that use unregistered IP addresses must use Network Address Translation to communicate with the rest of the world.

Here is an example of how NAT works.

An internal network (stub domain) has been set up with IP addresses that were not specifically allocated to that company by IANA (Internet Assigned Numbers Authority), the global authority that hands out IP addresses. These addresses should be considered non-routable since they are not unique.

The company sets up a NAT-enabled router. The router has a range of unique IP addresses given to the company by IANA.

A computer on the stub domain attempts to connect to a computer outside the network, such as a Web server.

The router receives the packet from the computer on the stub domain.

The router saves the computer’s non-routable IP address to an address translation table. The router replaces the sending computer’s non-routable IP address with the first available IP address out of the range of unique IP addresses. The translation table new has a mapping of the computer’s non-routable IP address matched with the one of the unique IP addresses.

When a packet comes back from the destination computer, the router checks the destination address on the packet. It then looks in the address translation table to see which computer on the stub domain the packet belongs to. It changes the destination address to the one saved in the address translation table and sends it to that computer. If it doesn’t find a match in the table, it drops the packet.

The computer receives the packet from the router. The process repeats as long as the computer is communicating with the external system.

What are ports and sockets?

A port is the means by which an application accesses the top layer of the TCP stack. Each application has its own individual port number. This ensures that data from one application does not get confused with data from another.

At the end of this document is a list of common port numbers and what application is associated with them.

As with the network and data link layers, one of the important functions of the transport layer protocol is to identify which application the data was created by and is destined for. TCP does this by specifying the port number assigned to the process that created the data. All applications are assigned a port number by the lANA (Internet Assigned Numbers Authority). When a TCP/IP packet arrives at its destination, the transport layer protocol receiving the IP datagram reads the value in the Destination Port field and delivers the information in the Data field to the program or protocol associated with that port.

When one TCP/IP system addresses traffic to another, it uses a combination of an IP address and a port number. The combination of an IP address and a port is called a socket. To specify a socket in a Uniform Resource Locator (URL), you enter the IP address first and then follow it with a colon and then the port number. So then the socket 192.168.2.10:21, for example, addresses port 21 on the system with the address 192.168.2.10. Since 21 is the port number for FTP, this socket addresses the FTP server running on that computer.

You usually don’t have to specify the port number when you’re typing a URL, because the program you use assumes that you want to connect to the well-known port. Your Web browser, for example, addresses all the URLs you enter to port 80, the Hypertext Transfer Protocol (HTTP) Web sewer port, unless you specify otherwise. The lANA port numbers are recommendations, not ironclad rules, however. You can configure a Web server to use a port number other than 80, and in fact, many Web servers assign alternate ports to their administrative controls, so that only users who know the correct port number can access them. You can create a semi-secret Website of your own by configuring your server to use port 81 (for example) instead of 80. Users would then have to type a URL like http://wwwmyserver.com:81 into their browsers instead of just http://www. myserver. com.

Appendix A

TOP enables the use of many different applications on a TCP/IP network. Below is a list of some of them.

Telnet

The Telnet program provides a remote login capability. This lets a user on one machine log onto another machine and act as though he or she were directly in front of the second machine. The connection can be anywhere on the local network or on another network anywhere in the world, as long as the user has permission to log onto the remote system.

File Transfer Protocol

File Transfer Protocol (FTP) enables a file on one system to be copied to another system. The user doesn’t actually log in as a full user to the machine he or she wants to access, as with Telnet, but instead uses the FTP program to enable access. Again, the correct permissions are necessary to provide access to the files.

Once the connection to a remote machine has been established, FTP enables you to copy one or more files to your machine. (The term transfer implies that the file is moved from one system to another but the original is not affected. Files are copied.) FTP is a widely used service on the Internet, as well as on many large LANs and WANs.

Simple Mail Transfer Protocol

Simple Mail Transfer Protocol (SMTP) is used for transferring electronic mail. SMTP is completely transparent to the user. Behind the scenes, SMTP connects to remote machines and transfers mail messages much like FTP transfers files. Users are almost never aware of SMTP working, and few system administrators have to bother with it. SMTP is a mostly trouble-free protocol and is in very wide use.

Kerberos

Kerberos is a widely supported security protocol. Kerberos uses a special application called an authentication server to validate passwords and encryption schemes. Kerberos is one of the more secure encryption systems used in communications and is quite common in UNIX.

Domain Name System

Domain Name System (DNS) enables a computer with a common name to be converted to a special network address. For example, a PC called Darkstar cannot be accessed by another machine on the same network (or any other connected network) unless some method of checking the local machine name and replacing the name with the machine’s hardware address is available. DNS provides a conversion from the common local name to the unique physical address of the device’s network connection.

Simple Network Management Protocol

Simple Network Management Protocol (SNMP) provides status messages and problem reports across a network to an administrator. SNMP uses User Datagram Protocol (UDP) as a transport mechanism. SNMP employs slightly different terms from TCP/IP, working with managers and agents instead of clients and servers (although they mean essentially the same thing). An agent provides information about a device, whereas a manager communicates across a network with agents.

Network File System

Network File System (NFS) is a set of protocols developed by Sun Microsystems to enable multiple machines to access each other’s directories transparently. They accomplish this by using a distributed file system scheme. NFS systems are common in large corporate environments, especially those that use UNIX workstations.

Remote Procedure Call

The Remote Procedure Call (RPC) protocol is a set of functions that enable an application to communicate with another machine (the server). It provides for programming functions, return codes, and predefined variables to support distributed computing.

Trivial File Transfer Protocol

Trivial File Transfer Protocol (TFTP) is a very simple, unsophisticated file transfer protocol that lacks security. it uses UDP as a transport. TFTP performs the same task as FTP, but uses a different transport protocol.

Transmission Control Protocol

Transmission Control Protocol (the TCP part of TCP/IP) is a communications protocol that provides reliable transfer of data. It is responsible for assembling data passed from higher-layer applications into standard packets and ensuring that the data is transferred correctly.

User Datagram Protocol

User Datagram Protocol (UDP) is a connectionless—oriented protocol, meaning that it does not provide for the retransmission of datagrams (unlike TCP, which is connection-oriented). UDP is not very reliable, but it does have specialized purposes. If the applications that use UDP have reliability checking built into them, the shortcomings of UDP are overcome.

Internet Protocol

Internet Protocol (IP) is responsible for moving the packets of data assembled by either TCP or UDP across networks. It uses a set of unique addresses for every device on the network to determine routing and destinations.

Internet Control Message Protocol

Internet Control Message Protocol (ICMP) is responsible for checking and generating messages on the status of devices on a network. It can be used to inform other devices of a failure in one particular machine. ICMP and IP usually work together.

Appendix B - Frequently used TCP port numbers

[table]

[table]

Appendix C — RFCs

The TCP/IP standards are published in documents called Requests for Comments (RFCs) by the Internet Engineering Task Force (IETF).

The list of RFCs contains documents that define protocol standards in various stages of development, but also contains informational, experimental, and historical documents that range from the fascinating to the downright silly. These documents are in the public domain and are accessible from many Internet Web and FTP sites. For links to the standards, see the IETF home page at [link]

Appendix D — Subnetting

Subnetting allows a single IP address (regardless of Class) to be divided into smaller pieces. This method of dividing addresses was introduced to help overcome some of the problems outlined above.

if the administrator of a corporate network wanted to create a new network at their site, they would have to request a new IP address for that network if it was to be accessible from public networks. Due to the finite number of addresses available this was not always possible. Subnetting adds another level of hierarchy to the IP addressing structure.

Instead of the classful two-level hierarchy (network number and host number), subnetting supports a three-level hierarchy:

[image]

Subnetting leaves the network prefix intact, but divides the host number into subnetwork prefix and host number. This allows a single network number to subdivided into smaller networks. Because the network prefix is the same, the route from the Internet to the host is the same, but once inside the network, routers will be used to determine which subnetwork the information is destined for. As far as the Internet is concerned all subnets belong to the same address, therefore only one entry in the routing table is required.

This practice allows local administrators to introduce varying levels of complexity to their private network without increasing the size of the Internet’s global routing table.

Design considerations

The deployment of an addressing plan requires careful thought on the part of the network administrator. There are four key questions that must be asked before any design should be undertaken:

1. How many total subnets does the organisation need today?

2. How many total subnets will the organisation need in the future?

3. How many hosts are there on the organisation’s largest subnet today?

4. How many hosts will there be on the organisation’s largest subnet in the future?

The first step in the planning process is to take the maximum number of subnets required and round up to the nearest power of two. For example if a network requires 9 subnets, 23 (=8) will not be enough subnet addressing space, so 2‘1 (=16) will be required.

The second step is to ensure that there are enough host addresses for the organisation’s largest subnet. If the largest subnet needs to support 50 host addresses, 25 (=32) will not provide enough addresses, so 26 (=64) will be required.

These calculations determine in what way the original IP address is divided to provide subnets. As with normal IP addresses, subnet addresses can vary in the amount of bits that identify the subnetwork number and number of bits that identify the host number.

As examined above, conventional IP addresses indicate what class of address they are using at the beginning on the address. Similarly subnet addresses indicate wht addressing scheme they are using (ie, where the divider between subnet number and host number lies). This is known as the subnet mask.

An example

Let’s assume that an organisation has been assigned the network number 193.1.1.0 (Class C) by the Internet’s governing body. The organisation decides it wishes to define 6 subnets. The largest subnet will need to support 25 hosts.

The first step is to determine the number of bits required to define the six subnets. Since a network address can only be subnetted along binary boundaries, subnets must be created in blocks of powers of two (2,4,8,16 etc). therefore it is not possible to create an IP address that defines 6 subnets (as 6 is not a power of 2). In this case the administrator must create 8 subnets and not use 2 of them (or have 2 reserved for future growth).

Now here’s where it gets complicated!

8 = 23. Therefore three bits will be required to enumerate the eight subnets.

The organisation has a class C address assigned to it, which means 24 bits are assigned to the network number. The new “extended” network prefix (network number and subnet number) will therefore be 27 bits long. The subnet mask will therefore be 255.255.255.224 to indicate that this is the addressing scheme used.

Don’t worry if that is as clear as mud! It makes more sense when you look at it in binary notation.

The organisation has been assigned the IP address 193.110. In binary notation this would be written:

11000001 .00000001 .00000001 .00000000

This is a class C address, which means the first three numbers define the network address, and the last number the host address on that network.

To create 8 subnets we are going to “borrow” three bits from the host address. To identify that this is what we are doing we create a subnet mask:

11111111.1111111.11111111.11100000

This defines the three hits of the host address that are being used to create an extended network prefix — now we only have 5 bits left to identify the host machine, but this is enough to identify up to 32 machines (25), so is enough for our purposes. In dotted notation this subnet mask is 255.255.255.224

With me so far? We’re not finished yet!

The 8 subnets will be numbered 0 through 7:

00000000

00000001

00000010

00000011

00000100

00000101

00000110

00000111

To identify the subnet, the administrator places the three-bit identifier (the last three bits in the bytes shown above) in the three-bit subnet identifier we have borrowed from the host number as shown above.

Therefore the eight subnets would be identified as follows:

Subnet #0 11000001 .00000001 .00000001 .00000000 193.1.1.0

Subnet #1 11000001.00000001.00000001.00100000 193.1132

Subnet #2 11000001 .00000001 .00000001 .01000000 19311.64

Subnet #3 11000001.00000001.00000001.01100000 19311.96

Subnet #4 11000001 .00000001 .00000001 .10000000 193.1.1.128

Subnet #5 11000001 .00000001 .00000001 .10100000 193.1 .1 .160

Subnet #6 11000001.00000001.00000001.11000000 193.1.1.192

Subnet #7 11000001 .00000001 .00000001.1 1 100000 193.1.1224

The remaining 5 bits can be used to identify individual host addresses — up to 32 in total.

Subnet conventions

It is not possible to use subnet masks that are either all 0s or all 1s as this can cause routing errors. Not all routers on the Internet have the ability to examine subnet information. Therefore an IP address of 193.1.1.0 would be indistinguishable from the same IP address using a subnet mask of 255.255.255.224:

[image]

These IP addresses are normally reserved for internal use anyway for broadcasting. If you want to send a message to all machines on the network you send it to 255.255.255.255 or 000.000.000.000 depending on how the router has been configured.

TCP rate adaption and error correction

This guide builds and develops on the areas and concepts covered in my Introduction to TCP/IP, and looks in more detail at the handshaking process that occurs between two modems when they agree on a mutually agreeable connection state.

A brief recap

You will remember that TCP is a transport layer protocol and provides an end-to-end communications channel between processes on each host system. The channel is reliable (has error-checking and correction capability), full-duplex and streaming. To achieve this functionality, TCP breaks up the data being passed to it from the session layer into separate segments and attaches a number of headers to the data containing information for the target system. An IP header is attached to this TCP “packet” containing addressing details of the target machine. This packet (or datagram or frame) is passed to the network layer for delivery.

TCP is said to be a unicast protocol as it supports the exchange of data between two parties only — it is not a multicast protocol.

TCP is a connection-oriented protocol in that it establishes a connection between the two parties in order to exchange data. It then ensures the reliable transfer of data by the acknowledging of all received packets. The protocol detects when segments of the data stream have been discarded, reordered, duplicated or corrupted. Where necessary it will retransmit damaged segments to the receiver. This implies that the sender must keep a local copy of all transmitted data until it has been successfully received. It is therefore also said to be reliable.

TCP is a full-duplex protocol in that it allows both parties to send and receive data within a single TCP connection.

TCP is a streaming protocol. Streaming implies that for every “write” action by the sender, there is a corresponding “read” action by the receiver. During the establishment of the connection, the sender and receiver agree on the Maximum Transmission Unit (MTU) — the maximum size of segment both parties can comfortably handle. This value is regular updated by both parties during the connection. Data is sent in a “stream” of packets.

TCP is also a rate adaptive protocol in that the rate of data transfer adapts to the network load conditions. There is no pre-defined TCP data transfer rate; if the network and the receiver increase in capacity after the connection has already been established, the sender will attempt to inject more data into the network to occupy this space. Similarly, if there is a sudden congestion, the sender will reduce the amount of data being put into the network to allow it to recover. This function aims to achieve the maximum amount of data transfer without incurring data loss.

The TCP Protocol Header

In the Introduction to TCP/IP I looked in some detail at the IP header used by the Internet Protocol. TCP also uses its own header structure — it is the values in this header which provide the error checking and correction capabilities of the protocol.

The TCP header structure uses a pair of 16-bit source and destination Port addresses. The next field is a 32-bit sequence number, which identifies the sequence number of the segment within the whole steam. The sequence number does not necessarily always start with a value of 1, the initial value is intended to prevent delayed data from an old connection inadvertently being incorrectly incorporated in a new connection.

After the sequence number is the Acknowledgement sequence number which is used to inform the originating end of the connection that data has been successfully received. A segment which is travelling to the sender to acknowledge receipt of a previously transmitted segment is known as an ACK.

Following the acknowledgement sequence number, comes the Data Offset field. This field contains six “flags” which contain information on the type of packet which is being sent. These flags are single bits and have a value of 1 or 0 to indicate on or off status.

the URG flag is used to indicate whether the packet is urgent or not

the ACK flag is used to indicate whether the packet is an ACK or not

the PSH flag is used if the sending system wants to “push” data to the remote application

the RST flag is used to reset the connection

the SYN flag (for synchronise) is used during the establishment of the connection

the FIN flag (for finish) is used to close the connection

following the data offset field comes the Window field which can contain an optional amount of redundant data, up to 16 bits) to ensure that the preceding fields all result in a header of the same size.

You then have the TCP Checksum which is used by the receiving machine to ensure that the received data has no errors.

Following the checksum are a number of TCP Options. As their name indicates these are optional settings, including:

Maximum receive segment size (MTU) — this option is used when the connection is opened. It is intended to inform the remote machine of the maximum segment size that the sender is willing to receive. This option is only used in the initial SYN packet (the initial packet exchange that opens a TCP connection).

Window scale — this option is used to address the issue of latency (the delay between transmission and reception of data across the network). This option sets the amount of user data the packet holds. In conditions of high delay, fewer packets containing more data may be preferable to a lot of packets containing a little data. The maximum amount of data the packet can hold must fall within the MTU value.

SACK-permitted — SACK stands for Selective Acknowledgment. Provided both parties support SACK this option can be enabled to reduce the amount of data sent over a congested network. Instead of acknowledging all received packets, the receiver only asks for the retransmission of packets it doesn’t receive (based on the segment numbers) or packets which are corrupted.

Connection establishment

The first phase of a TCP session is the establishment of the connection. This requires a three way handshake, ensuring that both sides of the connection have an unambiguous understanding of the parameters to be used in the connection. The operation can be described as follows:

1. The local system sends the remote system an initial sequence number using a SYN packet

2. The remote system responds with an ACK of the initial sequence number and the initial sequence number of the remote system response in a SYN packet

3. The local system responds with an ACK of this remote sequence number

The connection is then opened. In other words the local machine tells the other machine what sequence number its stream will begin with. The remote machine acknowledges receipt of this information, and tells the local machine what sequence number its acknowledgment packets will begin with. The local machine acknowledges receipt of this information.

Volume transfer

Once the connection has been established, TCP then endeavours to establish the point of dynamic equilibrium between maximum network efficiency and packet loss. This is why when you start to download a file over an Internet connection the download rate always starts off quickly, then steadily gets slower until it settles on a rate which may fluctuate a little. This equilibrium is very important. Increasing data transfer leads to network congestion which leads to data loss. Lost packets will require retransmission of data which will further increase congestion thus leading to a declining spiral. The speed of the connection will be that which guarantees the reliable delivery of data.

During the connection establishment, the receiving machine will advertise to the sender the size of its receiving buffer — this is the amount of data the receiving machine can store while data is being processed. The sending machine will ensure that the amount of data in transit over the network at any point does not exceed this value, ensuring that no data is being sent across the network that may be automatically discarded.

The sending machine must also keep in its send buffer a copy of all sent data until it has been successfully received by the target machine, in case it needs to be retransmitted. This buffer cannot be exceeded either. The size of the send and receive buffers on both machines are used to calculate what is called a “sliding window” to determine the amount of data that can be placed on the network at any time. This is known as the congestion window (cwnd). The size of the congestion window is the optimum amount of data that can be put on the network at any time. The point at which packet loss occurs as the congestion window grows too large is known as the threshold (the ssthresh value). In TCP parlance, therefore, the value of ssthresh should always be greater than the cwnd value. In the event of packet loss, both parties will lower the data throughput rate to the agreed minimum supported by both parties, then gradually increase it again until packet loss occurs again. This process continues until the session is closed. This is known as slow start. In TCP parlance, the size of the cwnd value is reduced, then gradually increased (one segment per RTT [round trip time] interval) until it exceeds the ssthresh value.

Packet loss

TCP can detect packet loss in two ways. First, if a single packet is lost within a sequence of packets, the receiving machine will generate duplicate ACKs for each packet that arrives after that packet, letting the sender know that it needs to resend the lost packet. Secondly, if a packet is lost at the end of the stream, there are no subsequent packets for the receiver to acknowledge with duplicate ACKs. But if the sender does not receive any ACK for that packet once its timer has expired it will assume packet loss and will retransmit that packet.

A single duplicate ACK is not a reliable indicator of packet loss. If a single packet has been routed slightly differently to the rest of the stream with a slightly higher delay it may arrive out of sequence. When the receiver receives the packet, it will acknowledge with an ACK to let the sender know it has received the packet. Three duplicate ACKs in succession will trigger the sender to resend the lost packet.

Another signal of packet loss is a complete cessation of ACK packets to the sender. The sender will not wait indefinitely for an ACK packet, after a predefined delay it will assume packet loss automatically and resend the packet(s). this delay is governed by the Retransmission timer. This timer is constantly refreshed to take into account differences in network load. If the retransmission timer triggers too early it will send duplicate information into the network unnecessarily. If it triggers too slowly it will slow down the flow of data.

Assisting TCP performance

Although I stated above that TCP is an end-to-end protocol which ensures the transmission of packets from host to target, it is possible for network elements in between to assist in optimising performance.

[image]

One function the network can employ is RED, or Random Early Detection. RED permits a network router to discard packets even if there is space on the network. While this may sound inefficient there is a good reason why this may be desirable.

RED uses a weighted average queue length to predict the likelihood of packet loss. In other words, as the queue length increases (the time it takes for the router to receive an ACK from the router it has forwarded a packet to), the likelihood of packet loss increases.

At this stage it is probably worth reminding you that packet loss is ultimately due to the TTL value (Time To Live) in the IP header. To avoid a situation where packets circle the network forever, all packets are set with a limit on the number of routers they can pass through before the packet is discarded automatically. As the packet passes through a router, the router decreases the TTL value in the IP header by 1. When a router receives a packet with a TTL value of 1 in the IP header it discards it automatically. If this concept seems unfamiliar read the Introduction to TCP/IP for more information.

RED aims to avoid a situation where all TCP streams experience congestion at the same time. A small flow of packets will be allowed to pass through, but a large stream of packets may be subjected to control if the network is already under heavy load. RED also aims to avoid a situation where there is a total loss of ACK packets reaching the sender resulting a connection termination, so it may favourably allow an ACK packet to reach it destination at the expense of another packet of “less” importance.

RED is a useful performance enhancement. Although TCP is an end-to-end protocol in that it uses error checking and packet loss detection measures to ensure the reliable transmission of data, it makes no assumptions at all about the status of the network. TCP uses packet loss as an indicator that it should slow data throughput and begin again gradually incrementing throughput until packet loss occurs again. Whilst all of the above sounds fiendishly clever, it is important to note that there is no real intelligence as such — a limit is reached, it slows down, and begins again and thus the process repeats itself.

It is important to note that RED will not necessarily discard the randomly selected packet. It is RED’s function to signal the sender that there is a potential for queue exhaustion. RED signals this likelihood of congestion by adding a Congestion Experienced (CE) flag in the Type of Service (ToS) field of the 1P header.

Conclusion

TCP is not a predictive protocol. It is an adaptive protocol that attempts to operate the network at the point of greatest efficiency.

Part 2

Encryption, Certificates and Authentication

Cryptography is a cornerstone of the modern electronic security technologies used today to protect valuable information — whether it is corporate information held on intranets, or whether it is credit card information being transferred across the Internet. The term cryptography is derived from the Greek kryptés meaning hidden, and graphein meaning to write.

Cryptography enables two users (the sender and the receiver) to convert information (or plaintext) into a cipher (scrambled text) by means of a key, The information is unreadable to anyone who is not in possession of this key. The information is then said to be encrypted. The sender of the information must share this key in a secure manner so only those who are authorised to view the information can do so.

Encryption not only guarantees privacy, but it also proves that the information originated from a known source - provided that the key has not been compromised. This guarantee of authenticity is important in today‘s reliance on secure data transfer. Anyone who gains access to the key can not only decrypt the cipher into plaintext and read the information, but could also use the key to impersonate the originator.

Systems of cryptography also include techniques and mechanisms for verifying that the originators of encrypted messages are authentic.

This guide will look at the reasons why encryption is used, the different technologies behind encryption and authentication, and will also look in some detail at the use of digital certificates in both of these processes.

It is important to note that no system is infallible. What cryptography seeks to do is not to provide perfect or risk-free security, but rather seeks to make the unauthorised access to data so complicated, so costly that the cost incurred in breaking security is more than the potential value that might be gained from the information accessed.

Security

Cryptography seeks to provide four basic functions:

Confidentiality — assurance that only authorised users can read or use confidential information, and that to anyone who does manage to eavesdrop on the transmission medium, the information will be unreadable.

Authentication — verification of the identity of the entities that communicate over the network.

Integrity — verification that the original contents of information have not been altered or corrupted, either deliberately or accidentally.

Nonrepudiation — assurance that a party involved in communication cannot falsely deny that a part of the communication actually occurred.

Mathematics to the rescue

To make information secure, the plaintext is converted to ciphertext by applying an encryption algorithm. The strength of the encryption depends on the strength of the algorithm.

An encryption algorithm is basically a very difficult mathematical problem — a problem so difficult that it is even difficult for a computer to resolve.

What does it mean for a mathematical problem to be difficult? To answer that let me first explain what is meant by an algorithm.

An algorithm explains the steps to take to resolve a mathematical problem. A simple algorithm which everyone knows is addition. Take the values a and b, apply the addition algorithm to give the output a+b

A problem is “difficult” if the fastest algorithm to solve the problem takes a long time relative to the input size.

Factoring 15 to give 3x5 is a simple problem. The problem is more difficult if the numbers involved are significantly larger but the algorithm remains the same. Addition and factorisation are example of polynomial algorithms — the time taken to resolve the problem is proportional to the size of the numbers involved.

Modern encryption methods rely on exponential algorithms — as the size of the numbers involved increases, the time taken to resolve the problem increases disproportionately.

To encrypt information, its value is adjusted by a pre-defined amount by applying an algorithm to it. The amount by which the information is adjusted is specified by a key. To decrypt the encrypted information the same key is required.

Types of encryption

There are two main categories into which an encryption system can fall:

Symmetric

Asymmetric (aka public-key)

Symmetric encryption involves using the same key for both encryption and decryption. Symmetric keys are also known as shared secret keys as the same key is shared by the sender and the receiver, but it is kept secret from any other parties. The diagram below illustrates a basic symmetric key encryption:

[image]

Symmetric encryption is used by TLS (Transport Layer Security, aka SSL) and IPSec. This method of encryption has drawbacks. Since the same key is used for both encryption and decryption, if anyone were to steal the key they would be able to decipher encoded messages.

Asymmetric, or public-key encryption uses different keys for the encryption and decryption of information. This requires the use of a public key (which is available to other entities on the network) and a private key (which is known only to the owner). The public key is used to encrypt information, but only the corresponding private key can be used to decrypt the information.

To use an analogy think of a strongbox with a double-acting deadbolt on it. The deadbolt can only be locked with one key. Once locked it cannot be unlocked with the same key. To unlock the box you require a second key. This key can only unlock the box, it cannot lock it. Neither key can be recreated from the other.

You take the locking key and copy it 1000 times and freely distribute it. If anyone wants to put anything in the strongbox they can lock it with their key. But you are the only one who can unlock it. The only drawback is that you cannot be sure who put the contents in the box as 1000 people have the same key. Security has been gained at the price of verification.

Asymmetric encryption is used by VPN (Virtual Private Network) solutions.

Public key encryption is considerably slower than symmetric key encryption due to the higher computational overhead — the processor has to do much more work to decrypt the information.

When talking about the efficiency of a cryptographic system, there are three distinct factors to take into account:

Computational overheads — how much computation is required to perform the public and private key transformations

Key size — how many bits are required to store the key pairs and any system parameters

Bandwidth — how many bits must be communicated to transfer an encrypted message or a signature

Secure Key Exchange

In order for symmetric key encryption to work, the secret key must be shared securely to prevent it falling into the wrong hands. Two of the most common key exchange algorithms are:

Diffie-Hellman Key Agreement

RSA Key Exchange Process

Both methods make it difficult for an intruder who intercepts network traffic to guess or calculate the key required to decrypt the information being transmitted. The Diffie-Hellman Agreement is largely accepted to be the more secure of the two.

Diffie-Hellman Key Agreement

This form of key exchange actually uses public-key encryption technology. Public key encryption was first proposed in 1975 by Stanford University researchers Whitfield Diffie and Martin Hellman.

The basic process is shown below:

[image]

To preserve security, the key itself is never transmitted. The client only needs to prove that it has the correct key. The procedure is as follows:

1. the client indicates to the server that it wants to connect.

2. the sewer then sends a random number (the challenge) to the client.

3. the client then performs a computation using its key and the random number, and sends the result (the response) back to the server.

4. the server performs the same computation using the same random number and its copy of the key.

5. if the keys match, the result of the computation will match, and the client will be authenticated and accepted.

Message Digests

In order to ensure that the decrypted information you receive has not been altered en route or become encrypted, message digest functions (aka hash functions) are used. These verify the integrity of the data. They are commonly 128 to 160 bits in length and provide a digital identifier for each file or document.

Message digest functions are mathematical functions that process information to generate a different value for each unique document. Identical documents have the same digest, but if even one bit in the document is changed the digest changes.

[image]

The two most common forms of message digest in use today are MD5 and SHA. SHA is considered the stronger of the two as it uses a longer key length — the longer the key length the higher the degree of security.

Digital Signatures

Just as a handwritten signature is used to identify an individual for legal proceedings or financial transactions, so digital signatures are used to identify electronic entities - to prove that the source from which the data is coming is really that which it claims to be, to provide authentication.

One possible method for creating a digital signature is to encrypt all data with the originator’s private key. But this is impractical for several reasons:

Ciphertext is considerably larger than plaintext and places excessive demands on network resources.

Public key encryption is slow and places a high computational load on processors. Transmitting large amounts of Ciphertext can be used by hackers to break the encryption.

The most common form of digital signature is to sign the message digest function with the originator’s private key. This produces only a small amount of Ciphertext, does not require large amounts of data to be transmitted over the network and does not burden the processor.

Who do you trust?

By themselves, private and public keys cannot provide proof that they belong to an alleged owner. There has to be a way to verify the identity of the owner of a key set. On a private network this is relatively easy. But for open networks such as the Internet trust relationships need to be verified by a third party who can provide assurance that key sets do correspond with certain entities.

This role is performed by Certification Authorities (CAs). A CA (such as VeriSign) issues digital certificates. These certificates are used to sign public and private encryption keys to prove that their identity is genuine.

If the CA is trusted, then any key that has been signed with a certificate issued by that CA is also trusted.

Digital certificates function similarly to identification cards such as passports:

It contains personal information to help trace the owner

It contains the information that is required to identify and contact the issuing authority

It is designed to be tamper resistant and difficult to counterfeit

It is issued by an authority that can revoke the certificate at any time

It can be checked for revocation by contacting the issuing authority

The current industry standard for digital certificates is X.509 version 3. The contents of a certificate is described in Appendix B.

Certificate servers

For use on an internal network you can use a certificate server and generate digital certificates internally. Windows 2000 server has a certificate server built in.

It is relatively easy to configure Internet Explorer to trust certificates issued by your own server. Open Internet Explorer. From the Tools menu select Internet Options.

Click on the Content tab

[image]

Click on the Certificates button

[image]

Click one of the following tabbed categories for the type of certificates you want to install or remove:

• Personal. Certificates in the Personal category have an associated private key. Information signed by using personal certificates is identified by the user’s private key data. By default, internet Explorer places all certificates that will identify the user (with a private key) in the Personal category.

• Other People. Certificates in the Other People category use public key cryptography to authenticate identity. based on a matching private key that is used to sign the information. By default, this category includes all certificates that are not in the Personal category (the user does not have a private key) and are not from CAs.

• Intermediate Certification Authorities. This category contains all certificates for CAs that are not root certificates.

• Trusted Root Certification Authorities. This category includes only self-signed certificates in the root store. When a CA’s root certificate is listed in this category, you are trusting content from sites, people, and publishers with credentials issued by the CA.

• Trusted Publishers. This category contains only certificates from trusted publishers whose content can be downloaded without user intervention, unless downloading active content is disabled in the settings for a specific security zone, Downloading active content is not enabled by default. For each available security zone, users can choose an appropriate set of ActiveX security preferences.

Depending on the version of SSL or TSL that the certificate was created using, you may need to configure Explorer’s security settings:

In the Internet Options window, click on the Advanced tab. Scroll down to the Security section and ensure that the necessary protocols are enabled:

[image]

Cryptanalysis attacks

Cryptanalysis is the practice of examining encrypted information to ascertain the key to decrypt it

— to break the code.

Symmetric key encryption is subject to key search attacks (also called brute force attacks). In these attacks, the attacker tries each possible key until the right key is found to decrypt the message. Most attacks are successful before all possible keys are tried.

In general, you can minimize the risk of key search attacks by choosing shorter key lifetimes and longer key lengths. A shorter key lifetime means that each key encrypts less information, which reduces the potential damage if one of the keys is compromised.

Longer key lengths decrease the possibility of successful attacks by increasing the number of combinations that are possible. For example, for a 40-bit key, there are 24° possible values. By using a personal computer that can try 1 million keys per second, an attacker can try all possible keys in about 13 days. However, an 128-bit key has 2‘25 possible values. If you could use a computer that would allow you to try 100 billion keys 3 second and you used 10 million of these computers, it would take about 1013 years to try every possible 128-bit key value.

Given a key of the same length, public key cryptography generally is more susceptible to attack than symmetric key cryptography, particularly to factoring attacks. In a factoring attack, the attacker tries all of the combinations of numbers that can be used with the algorithm to decrypt ciphertext. Factoring attacks are similar to keysearch attacks, but the number of possible factors varies with each algorithm and with the length of the public key and private key that are used. In general, for a given key length, a factoring attack on a public key requires fewer attempts to be successful than a key search attack on a symmetric key.

Neither of these two types of attack are used commonly. The most common form of attack is a plaintext attack. Hackers can collect amounts of ciphertext and compare it with plaintext to try to determine the encryption key in use. Standard email headers are sent as plaintext, for example.

Key Lifetimes

Key length is only one factor in the strength of both symmetric and asymmetric systems. The longer a key is used the more susceptible to attack it is. Keys should be changed at regular intervals and should always be changed immediately if there is any suspicion that the key has been compromised.

Appendix A - Authentication protocols

PAP (Password Authentication Protocol)

A clear—text authentication scheme used in Point-to-Point Protocol (PPP) connections Password Authentication Protocol (PAP) is not a secure form of authentication because the user’s credentials are passed over the link in unencrypted form. For this reason, Challenge Handshake Authentication Protocol (CHAP) or some other authentication protocol is preferable if the remote client supports it. If the password of a remote client using PAP has been compromised, the authentication server can be attacked using replay attacks or remote client impersonation.

How It Works

PAP uses a two-way handshake to perform authentication. Once the PPP link is established using the Link Control Protocol (LCP), the PPP client sends a username and password to the PPP server. The server uses its own authentication scheme and user database to authenticate the user, and if the authentication is successful, the server sends an acknowledgment to the client.

PAP is typically used only if the remote access server and the remote client cannot negotiate any higher form of authentication. The remote client initiates the PAP session when it attempts to connect to the PPP server or router. PAP merely identifies the client to the PPP server; the server then authenticates the client based on whatever authentication scheme and user database are implemented on the server.

EAP (Extensible Authentication Protocol)

The Extensible Authentication Protocol (EAP) is an extension to the Point-to-Point Protocol (PPP). EAP was developed in response to an increasing demand for remote access user authentication that uses other security devices. EAP provides a standard mechanism for support of additional authentication methods within PPP. By using EAP, support for a number of authentication schemes may be added, including token cards, one-time passwords, public key authentication using smart cards, certificates and others. EAP, in conjunction with strong EAP authentication methods, is a critical technology component for secure virtual private network (VPN) connections because it offers more security against brute-force or dictionary attacks and password guessing than other authentication methods, such as CHAP.

To find out if EAP authentication methods are being used, see your system administrator.

CHAP (Challenge Handshake Authentication Protocol)

An encrypted authentication scheme in which the unencrypted password is not transmitted over the network. CHAP is more secure than PAP because CHAP encrypts the transmitted password, while PAP does not. SPAP and MS-CHAP are vendor-specific implementations of the CHAP standard.

How It Works

A typical CHAP session during the PPP authentication process works something like this:

1. A client connects to a network access server (NAS) and requests authentication.

2. The sewer challenges the client by sending a session ID and an arbitrary string.

3. The client uses the M05 one-way hashing algorithm and sends the sewer the username, along with an encrypted form of the server’s challenge, session ID, and client password.

4. A session is established between the client and the server.

To guard against replay attacks, the challenge string is chosen arbitrarily for each authentication attempt. To protect against remote client impersonation, CHAP sends repeated, random interval challenges to the client to maintain the session.

MS CHAP (Microsoft’s implementation of CHAP)

MS CHAP negotiates a secure form of encrypted authentication by using Message Digest 5 (MD5), an industry-standard hashing scheme. CHAP uses challenge-response with one-way MD5 hashing on the response. In this way, you can prove to the server that you know the password without actually sending the password over the network.

By supporting CHAP and MD5, Windows 2000 is able to securely connect to almost all other PPP servers. When you connect to other remote access servers or clients, Windows 2000 remote access may negotiate plaintext authentication if the other product does not support encrypted authentication.

Where possible, MS-CHAP is consistent with standard CHAP. lts response packet is in a format specifically designed for Windows NT and Windows 2000, and Windows 95 and later, networking products. In addition, MS-CHAP does not require the use of plaintext or reversibly encrypted passwords.

A system administrator can define authentication retry and password changing rules for the users connecting to your server. A version of MS-CHAP is available specifically for connecting to a server running Windows 95. You must use this version if your connection is to a server running Windows 95.

Note

MS-CHAP v2 is a mutual authentication protocol, which means that both the client and the server prove their identities. If your connection is configured to use MS-CHAP v2 as its only authentication method, and the server that you are connecting to does not provide proof of its identity, your connection disconnects. Previously, servers could skip authentication and simply accept the call. This change ensures that you can configure a your connection can be configured to connect to the expected server.

MS CHAP 2

A new version of the Microsoft Challenge Handshake Authentication Protocol (MS-CHAP v2) is available. This new protocol provides mutual authentication, stronger initial data encryption keys, and different encryption keys for sending and receiving. To minimize the risk of password compromise during MS—CHAP exchanges, MS—CHAP v2 drops support for the MS—CHAP password change and does not transmit the encoded password.

For VPN connections, Windows 2000 Server offers MS-CHAP v2 before offering the legacy MS-CHAP. Windows 2000 dial-up and VPN connections can use MS-CHAP v2. Windows NT 4.0 and Windows 98 computers can use only MS-CHAP v2 authentication for VPN connections.

Kerberos

A method of securely authenticating users’ requests for access to services on a network. It was developed by the Massachusetts Institute of Technology (MIT), which based it on the Data Encryption Standard (DES).

Kerberos is the primary security protocol for Microsoft Windows 2000 domains and is used by domain controllers to verify the identity of the user and the integrity of data during a session. Kerberos has also been implemented on several UNIX platforms including OpenBSD.

How It Works

Kerberos uses a ticket-based method for granting a user access to a network service. When a Kerberos-enabled client wants to request a network service (such as network logon) from a Kerberos-enabled server, the client must first contact an authentication server (A8) to receive a ticket and an encryption key. The encryption key, called the session key, is used to unlock communication between the client and the server and thereby authenticate that communication. The initial ticket, often called the ticket-granting ticket (TGT), contains a copy of the session key and an identity, which is a randomly generated number. The AS passes the TGT and the identity back to the client, which stores the ticket in its ticket cache. When the client wants to access a particular service, it sends the ticket to a ticket-granting server (TGS). (The TGS and AS are usually the same machine.) The TGS gives the client a ticket that securely identifies the client to the service it is requesting. Finally, the client presents the ticket to the network service it is trying to access and is granted access to the resource as many times as desired until the ticket expires. When the client sends a ticket, the ticket is always accompanied by an authenticator message that is encrypted with the session key. This authenticator includes a time stamp, which is used to ensure that the ticket is legitimate.

In the Windows 2000 implementation of Kerberos. each domain controller has the Kerberos v5 services running on it, and a Kerberos client is built into each server and workstation running Windows 2000. The Kerberos services maintain encrypted user passwords and identities in Active Directory. When a user logs on to a domain controller, the initial Kerberos authentication enables the user to access available resources anywhere in the enterprise because authentication credentials issued by the Kerberos services of one domain are accepted by all domains within a domain tree or a domain forest.

The Kerberos service issues an initial ticket for the logon domain when a user logs on to a Windows 2000 workstation. Any server running Windows 2000 can then validate the client’s ticket without having to contact the domain Kerberos service. It can do this because servers running Windows 2000 share the encryption key that the Kerberos service uses to encrypt tickets. This encryption key is called the server key.

Appendix B — Description of Digital Certificates

[table]

Introduction to IP Version 6

Due to concerns over the depletion of the existing pool of IP addresses and the desire to provide additional functionality for modern devices, an upgrade of the current version of the Internet Protocol (IPv4) is in the process of standardisation — IP Next Generation (an9) or IP Version 6 (IPv6).

This document will explain the problems of IPv4 and how they are addressed in IPv6. I will also look at the new IPv6 header and the various signalling protocols used by routers to handle IPv6 traffic. ‘

Limitations of IPv4

IPv4 addresses have become relatively scarce, forcing some organisations to use NAT (Network Address Translation) to map multiple private addresses to a single public IP address. While NAT promotes the use of private address space it can interfere with certain higher layer protocols such as IPSec.

The increase in the number of Internet-connected devices ensures that the public IPv4 address space will eventually be depleted.

The growth in the size of the Internet has led to a corresponding increase in the size of Internet routing tables. There are routinely 85,000 routes in the routing tables of backbone Internet routers. As the size of routing tables increases so performance is decreased and journey times increase.

IP configuration either has to be done manually or with automatic configuration protocols such as DHCP (Dynamic Host Configuration Protocol). As more and more devices connect to the Internet configuration needs to be simpler.

The current implementation of IP has no support for security.

While IPv4 headers contain ToS (Type of Service) settings to prioritise data and ensure the delivery of data in real time, this has limited functionality.

Features of IPv6

New header format

The IPv6 header has a new format that is designed to keep overhead to a minimum. This is achieved by moving non-essential and optional fields to extension headers that are placed after the IPv6 header. This means headers are more efficiently processed by routers. IPv4 and IPv6 headers are not interoperable. A host or router must use an implementation of both IPv4 and IPv6 in order to recognise and process both header formats. IPv6 addresses are four times as large as IPv4 addresses. but the header is only twice as large.

Large address space

IPv6 has 128-bit source and destination IP addresses. 128 bits can express 3.4 x 1035 possible combinations. This large address space allows for multiple levels of subnetting and means that individual subnets within an organisation can be assigned an IP address from the Internet authority, thus techniques such as NAT will no longer be necessary.

Efficient and hierarchical addressing and routing infrastructure

IPv6 addresses are designed to create an efficient and hierarchical routing infrastructure meaning that Internet backbone routers have smaller routing tables.

Stateless and stateful address configuration

To simplify host configuration, IPv6 supports both stateful configuration (in the presence of a DHCP server) and stateless configuration (in the absence of a DHCP server). Stateless hosts configure themselves for the duration of the configuration using network prefixes advertised by the router. Even in the absence of a router hosts can configure themselves automatically without manual configuration.

Built-in security

IPv6 supports optional packet encryption using iPSec.

Better support for 008

New fields in the IPv6 header define how traffic is handled and identified. Traffic identification using a Flow Label field in the header allows router to identify and provide special handling for packets belonging to a flow (a series of packets between a source and a destination). Because this information is contained in the header, QoS settings can be used even if the payload of the packet is encrypted.

New protocol for neighbouring node interaction

The Neighbour Discovery protocol is a new ICMP (Internet Control Message Protocol) message that manages the interaction of neighbouring nodes on the same link. Neighbour Discovery replaces the broadcast-based Address Resolution Protocol (ARP).

Extensibility

IPv6 can be easily extended for new features by adding extension headers after the IPv6 header. Unlike options in the IPv4 header which is limited to 40 bytes, the size of the IPv6 extension headers is only constrained by the size of the IPv6 packet.

IPv6 Addressing

The most obvious distinguishing feature of IPv6 is its use of much larger addresses. The size of an address is 128 bits - four times larger than the IPv4 address. A 32-bit address space allows for 232 or 4,294,967,296 possible addresses. A 128-bit address space allows for 2^28 addresses (340,282,366,920,938,463,463,374,607,431,768,211,456). It is hard to imagine a scenario where this many addresses would not be enough!

Syntax

IPv4 addresses are represented in dotted-decimal format. This 32-bit address is divided along 8-bit boundaries. Each set of 8 bits is converted to its decimal equivalent and separated by dots. For IPv6, the 128-bit address is divided along 16-bit boundaries, and each 16-bit block is converted to a 4-digit hexadecimal number and separated by colons. The resulting representation is called colon-hexadecimal:

21DA:D3:O:2F3B:2AA:FF:FE28:905A

Some types of addresses contain long sequences of zeros. To further simplify the representation of IPv6 addresses, a 16-bit sequence containing all zeros is omitted.

IPv6 addresses are separated into network prefix and host number in the same way as IPv4 addresses.

There are three types of IPv6 address:

Unicast - a unicast address identifies a single interface on the network

Multicast — a multicast address identifies multiple interfaces on the network

Anycast — an anycast address identifies all of the interfaces on the immediate network (immediacy being determined by routing distance)

Aggre gatable Global Unicast Addresses

As the name implies, Aggregatable global unicast addresses are designed to be summarised to produce an efficient routing hierarchy. A router running IPv6 only needs to look at the first few bits of the target address to know where to send the packet. This reduces the amount of processing done by the router, reduces delay and therefore overall data throughput.

The structure of an aggregatable unicast address is as follows:

[image]

TLA ID

The TOP-Level Aggregation Identifier is a 13-bit field. This identifies the highest level in the routing hierarchy. These IDs are administered by the Internet authority and allocated to local lSPs. This field size allows for up to 8,192 TLA le.

Res

Bits that are reserved for future use I expanding the size of either the TLA ID or the NLA ID.

NLA ID

The Next-Level Aggregation Identifier is used to identify a specific customer site. The field is 24- bits. Together the TLA ID and the NLA ID form the public part of the IP address.

SLA ID

The Site-Level Aggregation Identifier is used by an individual organisation to identify subnets within its site. The field is 16-bits, allowing for the creation of up to 65,536 subnets or multiple levels of addressing hierarchy.

Interface ID

The Interface ID identifies the interface (network card) on a specific subnet.

IPv6 Header

The IPv6 header is a fixed size of 40 bytes:

[image]

Before we look at the changes in header structure in IPv6, let’s look briefly at the current header structure used in IPv4

[image]

Version Indicates the version of IP used

Internet Header Length Indicates the total length of the header. The minimum size of the IPv4 header is 20 bytes. Options can extend this by a number of additional 4-byte blocks, with padding being used if necessary to fill unused space.

Type of Service Indicates the desired service expected by this packet for delivery through routers across the network. This field is 8—bits and contains parameters for precedence, delay, throughput and reliability.

Total Length Indicates the total length of the whole packet (header and payload). The size of this field is 16-bits, which can indicate a maximum of 65,535 bytes.

Identification Identifies the specific packet in a stream of packets. If a packet is fragmented during transmission, all of the fragments retain the original identification value so that they can be grouped for reassembly.

Flags Identifies flags for the fragmentation process. The size of this field is 3 bits, however, only 2 bits are defined for use. There are two flags — one to indicate whether the packet might be fragmented and another to indicate whether more fragments follow the current fragment.

Fragment Offset Indicates the position of the fragment relative to the original packet. ‘

Time to Live Indicates the maximum number of links on which a packet can travel before being discarded. As a packet passes through a router, the TTL value is decreased by one. When a router receives a packet with a TTL value of 1 it is discarded and an ICMP Time

Expired message is sent back to the sending node.

Protocol Identifies the upper layer protocol. The size of field is 8 bits. For example, TCP uses a protocol of 6, UDP 17 and ICMP 1.

Header Checksum Provides a checksum on the IP header. The size of this field is 16 bits. The payload is not included in the checksum. Each node that receives the packet verifies checksum integrity and discards the packet if verification fails. Because the router must decrease the TTL value, the header checksum is recalculated with each hop.

Source Address Stores the IP address of the originating host. Destination Address Stores the IP address of the destination host. Options Stores one or more lP options. The size of this field is a multiple of 32 bits. If the IP option or options do not use all 32 bits, padding is used so that the IP header is a number of 4-byte blocks that can be indicated by the Internet Header Length field.

The IPv6 Header:

[image]

Version 4 bits are used to indicate the version of IP used (set to 6)

Traffic Class Indicates the class or priority of the packet. This field provides similar functionality to the va4 Type of Service field.

Flow Label Indicates that this packet belongs to a specific sequence of packets between a source and destination, requiring special handling by intermediate routers. The Flow Label is used for non-default quality of service connections, suCh as those needed by real—time data (voice and video). For default router handling, the Flow Label is set to O.

Payload Length Indicates the length of the total packet. The size of this field is 16 bits providing a maximum of 65,535 bytes. For payload lengths greater than 65,535 bytes the Payload Length field is set to 0 and the Jumbo Payload option is used in the Extensions header.

Next Header Indicates either the first extension header (if present) or the protocol in the upper layer (such as TCP, UDP or lCMP). The size of the field is 8 bits. When indicating an upper layer protocol, the same values as IPv4 are used.

Hop Limit Indicates the maximum number of links over which the packet can travel before being discarded. The Hop Limit is similar to the IPv4 TTL field.

Source Address Stores the va6 address of the originating host.

Destination Address Stores the va6 address of the destination host. If a routing extension header is present, the Destination Address might be set to the next router interface.

Extension Headers

ln IPv4 the IP header contains all of the options. As a packet passes through a router, all of these options must be processed, causing performance degradation. With IPv6 all options have been moved to extension headers. The only extension header that must be processed by the router is the Hop-by—Hop Options extension header. This increases processing speed and forwarding performance.

Currently these are the extension headers supported by va6:

Hop-by-Hop Options

Destination Options

Routing header

Fragment header

Authentication header

Encapsulating Security Payload (ESP) header

Extension headers are processed in the order in which they are present, therefore there are rules to the order in which they can be placed. Because the Hop-by-Hop Options extension header must be processed, it must be placed first.

Extension headers must fall on a 64-bit (8 byte) boundary Extension headers of variable length must use padding as required to ensure this.

The Next Header field in the IP header indicates whether an extension is present or not. Each extension header also has its own Next Header value to indicate whether another extension is present:

[image]

Hop-by-Hop Options header

The Hop-by-Hop Options header consists of a Next Header field, a Header Extension Length field, and an Options field that contains one or more options.

The options contain information on how the packet should be handled by the router: The Jumbo Payload option is used to indicate a payload greater than 65,535 bytes. The option is 32-bits, therefore with the Jumbo Payload option, payload sizes of up to 4,294,967,295 bytes can be indicated. An IPv6 packet with a payload greater than 65,535 bytes is known as a Jumbogram.

The Router Alert option is used to indicate to the router that the contents of the packet require additional processing.

Destination Options header

The Destination Options header is used to specify packet delivery parameters for either intermediate destinations or the final destination. It is used in two ways:

If a Routing header is present, it specifies delivery or processing options at each intermediate destination It specifies delivery or processing options at the final destination

Routing header

Much like va4, va6 source nodes can use the Routing header to specify loose routing. Alternatively specific routing information with the addresses of each intermediate host can be included. As the packet arrives at each intermediate host on the route, the Routing header is processed and the address of the next intermediate host becomes the destination address in the IP header.

Fragment header

The Fragment header is used for fragmentation and reassembly services. In va6, only source nodes can fragment payloads. If the payload submitted by the upper layer protocol is larger than the link MTU (Maximum Transmission Unit), the IP fragments the payload at the source and includes reassembly information. When an va6 packet is fragmented, it is initially divided into unfragmentable and fragmentable parts:

The unfragmentable part of the original IPv6 packet must be processed by each intermediate node between source and destination. This part therefore contains the IP header, Hop-by-Hop Options header, Destination Options header and the Routing header.

The fragmentable part of the original IPv6 packet must only be processed by the final destination. This part consists of the Authentication header, the ESP header, the Destination Options header for the final destination and the upper layer PDU.

The fragmentation process is illustrated below:

[image]

Authentication header

(see Jimbo’s Guide to Encryption, Authentication and Digital Certificates for more information on this subject)

The Authentication header provides data authentication (verification of the node that sent the packet), data integrity (verification that the data was not modified in transit — either intentionally or unintentionally) and anti-relay protection (assurance that captured packets cannot be retransmitted and accepted as valid). The Authentication header (AH) is part of the IPv6 security architecture and is already used by VPN protocols such as IPSec.

The AH does not provide data confidentiality by encrypting the contents of the payload. To do this ‘ it can be used in conjunction with the Encapsulating Security Payload (ESP). See Jimbo’s Guide to Virtual Private Networking for more information on this subject.

Encapsulating Security Payload header and trailer

The ESP header and trailer provide data confidentiality, data authentication and data integrity services. The ESP header contains the Security Parameters index (SPI) field that identifies the iPSec session information, as well as a Sequence Number field that provides anti-relay protection.

Details on how both the AH and the ESP work are beyond the scope of this document.

Signalling

IPv6 uses a revised version of the Internet Control Messaging Protocol (ICMPv6). There are two types of ICMP message:

Error messages

Informational messages

Error messages are used to report errors in the forwarding or delivery of lP packets by either the destination node or an intermediate node. Error messages include Destination Unreachable, Packet Too Big, Time exceeded and Parameter Problem.

Informational messages are used to provide diagnostic functions and additional host functionality.

Destination Unreachable messages

Destination Unreachable messages can have a value between 0 and 4:

0 No route matching the destination was found in the table

1 The communication with the destination is prohibited by administrative policy. (This

is typically sent when the packet is blocked by a firewall)

2 The address is beyond the scope of the source address

3 The destination address is unreachable. This is typically sent because of an

inability to resolve the destination’s link layer address

4 The destination port was unreachable. This is typically sent when a packet containing a UDP message arrived at the destination but there were no applications listening on the destination UDP port

Packet Too Big error message

An ICMP Packet Too Big message is sent when the packet cannot be forwarded because the link MTU is smaller than the size of the packet. IPv6 requires that the link layer support a minimum packet size of 1280 bytes. Link layers that do not support this must provide a link layer fragmentation and reassembly scheme that is transparent to IP.

Time Exceeded error message

An ICMP Time Exceeded message is sent by a router when the Hop Limit field in the IP header is zero.

Parameter Problem error message

An ICMP Parameter Problem message is sent either by a router or the destination host. This occurs when an error is encountered in either the IP header or an extension header, preventing IP from performing further processing.

Informational messages provide diagnostic capabilities to aid troubleshooting.

Echo Request

An ICMP Echo Request message is sent to a destination to solicit an immediate Echo Reply message. The Echo Request/ Echo Reply message facility provides a simple diagnostics function to aid in the troubleshooting of a variety of reachability and routing problems.

Path MTU Discovery

The path MTU (Maximum Transmission Unit) is the lowest common denominator in any IP link. Packets with a maximum size of the path MTU do not need to be fragmented and will be successfully forwarded by all routers on the path. To discover the path MTU, the sending node uses the receipt of ICMP Packet Too Big messages.

The path MTU is discovered through the following process:

1. The sending node assumes that the path MTU is the link MTU of the interface on which the traffic is being forwarded. (In other words lP knows the MTU of the network segment it is on, and assumes that this is the MTU of the entire path to the destination)

2. The sending node sends IP packets at the path MTU size

3. If a router on the path is unable to forward the packet over a link with an MTU that is smaller than the size of the packet. it discards the packet and sends an iCMP Packet Too Big message back to the sending node. The lCMP message contains the MTU of the link on which the forwarding failed.

4. The sending node sets the path MTU for packets being sent to the destination to the value of the MTU in the lCMP message.

This process may need to be repeated several times.

Multicast Listener Query

MLQ is a service used by routers to establish what interfaces are to be addressed by multicast messages.

Neighbour Discovery

Neighbour Discovery (ND) is a set of messages used by both hosts and routers. It is used by hosts to:

Discover neighbouring routers

Discover addresses, address prefixes, and other configuration parameters

ND is used by routers to:

Advertise their presence, host configuration parameters, and on-link prefixes

Inform hosts of a better next-hop address to forward packets for a specific destination

ND is used by nodes to:

Resolve the link layer address of a neighbouring node to which an IP packet is being

forwarded and determine when the link layer address of a neighbouring node has

Changed

Determine whether a neighbour is still reachable

Address Autoconfiguration

One of the most useful aspects of IPv6 is its ability to configure itself, even without the use of a

stateful configuration protocol such as DHCP. By default, an IPv6 host can configure a link address for each interface. By using router discovery, a host can also determine the address of routers, other configuration parameters, additional addresses and link prefixes.

The address autoconfiguration process for the physical interface of an IPv6 node is as follows:

1. A tentative link address is derived based on the link prefix

2. Using duplicate address detection to verify the uniqueness of the tentative link address,a

Neighbour Solicitation message is sent with the Target Address field that is set to the tentative link address

1. If a Neighbour Advertisement message sent in response to the Neighbour Solicitation message is received, this indicates that another node on the local link is using the tentative link address. At this point the process stops and manual configuration must be done on the node.

2. If no Advertisement message is received, the tentative link address is assumed to be unique and valid. The local address is therefore initialised for the interface.

The next step of the configuration process then continues:

1. The host sends up to 3 Router Solicitation messages (by default)

2. if no router Advertisement messages are received, then the host uses a stateful address configuration protocol to obtain addresses and other configuration parameters

3. If a Router Advertisement message is received, the Hop Limit, Reachable Time, Retrans Timer and the MTU are set

4. 4. Duplicate address detection is then used to verify the uniqueness of the Tentative address

The process is illustrated below:

[image]

Voice over IP (VOIP)

VOIP is the transmission of voice calls over data networks such as the Internet. This technology has many benefits — imagine being able to make a long—distance call to the other side of the world for the price of a call to a local ISP. Not many people know that VOIP is already used by many telephone companies to handle their long-distance communications between regional offices.

The conventional PSTN (Public Switched Telephone Network) relies on circuit switching. Connections are established between the two parties, and the connection remains open until one party closes it by hanging up. A typical telephone call may happen as follows:

1. You pick up the receiver and listen for a dial tone. This lets you know that

you have a connection to the local office of your telephone carrier.

1. You dial the number of the party you wish to talk to.

2. The call is routed through the switch at your local carrier to the party you are calling.

3. A connection is made between your telephone and the other party’s line, opening the circuit.

4. You talk for a period of time and then hang up the receiver.

5. When you hang up, the circuit is closed, freeing your line.

Let’s say that you talk for 10 minutes. During this time, the circuit is continuously open between the two phones. Telephone conversations over the traditional PSTN are transmitted at a fixed rate of about 64 kilobits per second (Kbps), or 1,024 bits per second (bps), in each direction, for a total transmission rate of 128 Kbps. Since there are 8 kilobits (Kb) in a kilobyte (KB), this translates to a transmission of 16 KB each second the circuit is open, and 960 KB every minute

it’s open. 80 in a 10-minute conversation, the total transmission is 9600 KB, which is roughly equal to 9.4 megabytes (MB).

Because during a typical telephone conversation only one party is speaking at a time, half of this transmitted data is wasted. If we could remove this “dead air”, then the transmitted data could be cut down to 4.7 MB

Data networks do not use circuit switched connections. When you retrieve a web page, the connection with the web server is dropped once the page has been successfully transferred. Instead, data networks use packet switching.

Note — if you are making a dial—up connection to the internet, you still pay for the time you are connected and in this sense the connection is circuit-switched, but the connection is only maintained with the lSP’s portal to the Internet, not with the web servers you may connect to subsequently.

Packet switching is very efficient: it minimises the time that a connection is maintained between two systems, which reduces the load on the network. It also frees up the two computers communicating with each other so that they can accept information from other computers as well.

VOIP uses this technology to provide several advantages over traditional circuit- switching. Using PSTN, that 10-minute phone call consumed 10 full minutes of transmission time at a cost of 128 Kbps. With VOIP, that same call may have occupied only 3.5 minutes of transmission time at a cost of 64 Kbps, leaving another 64 Kbps free for that 3.5 minutes, plus an additional 128 Kbps for the remaining 6.5 minutes. Based on this simple estimate, another three or four calls could easily fit into the space used by a single call under the conventional system. And this example doesn’t even factor in the use of data compression, which further reduces the size of each call.

At a basic level, IP telephony and conferencing technologies are built on simple concepts. A personal computer (or other device) is used to capture audio and video signals from the user (for example, by using a microphone attached to a sound card, and a video camera attached to a video capture card). This information is compressed and sent to the intended recipient(s) over the LAN or the Internet. At the receiving end, the signals are restored to their original form and played back.

On a more complicated level, using VOIP, a company’s PBX becomes a gateway to the IP network (the internet, for example). in this system a typical call may happen as follows:

1. You pick up the receiver, which sends a signal to the PBX.

2. The PBX receives the signal and sends a dial tone. This lets you know that you have a connection to the PBX.

3. You dial the number of the party you wish to talk to. This number is then temporarily stored by the PBX.

4. Once you have entered the number, the PBX checks it to ensure that it is in a valid format.

5. The PBX determines whom to map the number to. in mapping, a telephone number is mapped to the lP address of another device called the IP host. The IP host is typically another digital PBX that is connected directly to the phone system of the number you dialled. In some cases, particularly if the party you are calling is using a computer-based VOIP client, the lP host is the system you wish to connect with.

6. A session is established between your company’s PBX and the other party‘s IP host. This means that each system knows to expect packets of data from the other system. Each system must use the same protocol to communicate. The systems will implement two channels, one for each direction, as part of the session.

7. You talk for a period of time. During the conversation, your company’s PBX and the other party’s IP host transmit packets back and forth when there is data to be sent. The PBX at your end keeps the circuit open between itself and your phone extension while it forwards packets to and from the IP host at the other end.

8. You finish talking and hang up the receiver.

9. When you hang up, the circuit is closed between your phone and the PBX, freeing your line.

10. The PBX sends a signal to the IP host of the party you called that it is terminating the session.

11. The IP host terminates the session at its end, too.

12. 0nce the session is terminated, the PBX removes the number-to-lP-host mapping from memory.

In order for telephone systems to be able to communicate with each other a data network they need to share a common protocol.

VOIP Protocols

VOIP currently uses two main protocols — H232 and SIP. Both protocols are responsible for three main areas of communication: establishing the call, handling changes to the call and ending the call. These responsibilities fall into two categories: signalling and session management. Signalling allows information to be carried over the network. Session management provides the ability to control the attributes of a call and also define codecs (compressor-decompressor) which are used to convert audio signals into a compressed digital signal for transmission and back again for audio replay.

H.232

The first protocol is H.323. H232 originated as an ITU (International Telecommunications Union) protocol and is actually a suite of protocols which provide support for a number of specific applications such as videoconferencing, data sharing and IP telephony. It was intended as an “all-in-one” solution to ensure that systems from different vendors worked together. However it is also relatively slow and cumbersome due to the large amount of redundant data transmitted. It is also very inflexible in terms of upgrades. In a field as untested and ever-changing as IP telephony a more flexible protocol is required.

I won’t look at this protocol in any detail, but more information can be found at

http://www.protocols.com

Configuring H323 settings on Windows XP

Open the Control Panel

Double click on Phone and Modem Options

Click on the Advanced tab

[image]

Select Microsoft H.323 Telephony Service Provider and click Configure

[image]

Here you can specify whether you wish to use a gateway or proxy server and enter the name of the server. You can also specify what port number to wait or incoming calls on.

An H.323 gateway connects the IP network with a switched service network, such as the public switched network (PSTN). A proxy server acts as a firewall or security barrier between your intranet and the Internet, keeping other people on the Internet from gaining access to confidential information on your internal network or your computer. Your telephony system administrator can provide the correct name or IP address to enter here.

Initiating a VOIP call

You can use the Windows Phone Dialer application to initiate a VOIP call.

You can run Phone Dialer by typing “dialer.exe” in the Run dialogue box on the

Start Menu:

[image]

To initiate the call open the Phone menu and select Dial

[image]

Select the option to make an Internet call and enter the recipient’s IP address:

[image]

If you will be calling this number on a regular basis you can select the option to add the number to your speed dial list

Alternatively you can save a contact’s IP address in the Windows Address Book.

Open the Address Book and create a new contact. Click on the Business tab.

You will see a field for IP Phone:

[image]

Receiving VOIP calls

In order to receive a VOIP call the Phone Dialer application must be running. It can be minimised to the system tray for convenience.

[image]

Session Initiation Protocol (SIP)

The second protocol, a potential successor to H.323, is the Session Initiation Protocol (SIP).

Currently VOIP does not offer sufficient voice quality to be seriously considered as a viable alternative to more conventional telephony solutions, like the PSTN.

The Internet Engineering Task Force (IETF) is committed to making high-quality voice over IP a reality. The Session Initiation Protocol is at the heart of this project, and is modelled on other well-established Internet protocols such as SMTP (Simple Mail Transfer Protocol) and HTTP (Hypertext Transfer Protocol). It reuses existing technology not only to make it interoperable with existing protocols such as email, but also to provide easy extensibility in the future.

SIP has a number of advantages over H232. It has the ability to:

Determine the location of the target endpoint—SIP supports address resolution, name mapping, and call redirection.

Determine the media capabilities of the target endpoint—Via Session Description Protocol (SDP), SIP determines the “lowest level” of common services between the endpoints. Conferences are established using only the media capabilities that can be supported by all endpoints.

Determine the availability of the target endpoint—If a call cannot be completed because the target end point is unavailable, SIP determines whether the called party is already on the phone or did not answer in the allotted number of rings. It then returns a message indicating why the target endpoint was unavailable.

Establish a session between the originating and target endpoint—If the call can be completed, SIP establishes a session between the endpoints. SIP also supports mid-call changes, such as the addition of another endpoint to the conference or the changing of a media characteristic or codec.

Handle the transfer and termination of calls—SIP supports the transfer of calls from one end point to another. During a call transfer, SIP simply establishes a session between the transferee and a new endpoint (specified by the transferring party) and terminates the session between the transferee and the transferring party. At the end of a call, SIP terminates the sessions between all parties.

SIP provides support for the following call features:

Call Hold

Consultation Hold

Unattended Transfer

Call Fwd Unconditional

Call Fwd on Busy

Call Fwd on No Answer

3-Way Conference

Single Line Extension

Find-Me

Incoming Call Screening

Outgoing Call Screening

Secondary Number — In

Secondary Number — Out

Do Not Disturb

Call Waiting

SIP Protocols

SIP is actually a suite of protocols, much in the same way that TCP is actually comprised of several protocols all working together.

SIP has two main components: the SIP User Agent and the SIP Network Sewer. The User Agent can be a lightweight piece of software, suitable for embedding in a PDA or Smartphone device. The main function of the SIP Server is to provide name resolution and location information.

The User Agent is effectively the end system component for the call and the SIP Server is the network device that handles the signalling associated with multiple calls. The User Agent itself has a client element, the User Agent Client (UAC) and a server element, the User Agent Server (UAS.) The client element initiates the calls and the server element answers the calls. This allows peer-to-peer calls to be made using a client-server protocol.

The SIP Server element also provides for more than one type of server. There are effectively three forms of server that can exist in the network — the SIP stateful proxy server, the SIP stateless proxy server and the SIP redirect server. The main function of the SIP sewers is to provide name resolution and user location, since the caller is unlikely to know the IP address or host name of the called party. What will be available is perhaps an email—like address or a

telephone number associated with the called party. Using this information, the caller’s user agent can identify with a specific server to “resolve” the address information. It is likely that this will involve many servers in the network. In this respect, identity resolution will be similar to the DNS system. Indeed SIP borrows substantially from SMTP and its routing system.

A SIP proxy server receives requests, determines where to send these, and passes them onto the next server (using next hop routing principles). There can be many server hops in the network.

The difference between a stateful and stateless proxy server is that a stateful proxy server remembers the incoming requests it receives, along with the responses it sends back and the outgoing requests it sends on. A stateless proxy server forgets all information once it has sent on a request. This allows a stateful proxy server to fork requests to try multiple possible user locations in parallel and only send the best responses back (in other words, if there are multiple server entries for the same user, it will try all of them and then use the address which the user is actually using). Stateless proxy servers are most likely to be the fast, backbone of the SIP infrastructure. Stateful proxy servers are then most likely to be the local devices close to the User Agents, controlling domains of users and becoming the prime platform for the application services.

A redirect server receives requests, but rather than passing these onto the next server it sends a response to the caller indicating the address for the called user. This provides the address for the caller to contact the called party at the next server directly.

Networks can also implement a SIP Registrar. This device accepts registration requests from users and maintains a Location Server database much like the

HLR in a GSM network.

The Session Description Protocol is responsible for negotiating a mutually— acceptable call framework between the two parties. It decides the media to use (codec and sampling rate), the media destination (IP address and port number), the session name and the time the session is active for.

SIP and SDP are signalling protocols that enable one party to place a call to another party and to negotiate the parameters of a multimedia session. The actual audio, video or other multimedia content is exchanged between session participants using an appropriate transport protocol. In many cases, the transport protocol to use is the Real-time Transport Protocol (RTP). RTP is used to packetise various digital media streams, such as voice, text and video.

Real Time Streaming Protocol is responsible for ensuring that packets in real time media streams arrive consecutively.

How does it work?

SIP is a simple, ASCII-based protocol that uses requests and responses to establish communication among the various components in the network and to ultimately establish a conference between two or more endpoints.

A SIP request message is comprised of three elements:

Request line

Header

Message body

A SIP response message also consists of three elements:

Status line

Header

Message body

Users in a SIP network are identified by unique SIP addresses. A SIP address is similar to an email address and is in the format of sip:userlD@gateway.com. This address is also known as a Universal Resource Indicator (URI).

When a user initiates a call, a SIP request is sent to a SIP server (either a proxy or a redirect server) by the SIP User Agent. The request includes the address of the caller (in the From header field) and the address of the intended callee (in the To header field).

The SIP server will then try to resolve the address, much in the same way that a web server or email address is resolved by DNS. The location of the end user will be registered with the SIP server on the target user’s network. Because over time a user may move between end systems, the server may have multiple addresses for the user. If a Stateful Proxy SIP Server is being used, this server will try each of the addresses in turn until it locates the user, and return this address. If a Stateless Proxy Server or a Redirect Server is being used, all of the addresses will be returned in the header of the response message.

The SIP server then returns this information to the SIP User Agent. The User Agent then sends an INVITE request to the server, which forwards it to the callee. The caller and callee then exchange messages indicating their capabilities in terms of available services, transmission speed and quality of service*. The requests and responses involved in these exchanges are forwarded by the server. Once a mutually-acceptable framework has been agreed upon, a session is opened between the two parties.

* SIP does not support quality of service. A minimum required service level can be specified and the call will only proceed if this level can be guaranteed.

Interoperability

H232 and SIP are not directly compatible with each other, but both protocols can exist in the same network if a device that supports both protocols is available. For example, a User Agent could communicate with the Server via H232, and the server could then communicate with other parties via SIP.

Usage scenarios

Voice calls can be made over an IP-based network in a number of ways:

From PC to PC

From PC to PSTN phone

From PSTN phone to PC

From IP phone to IP phone

Alternative uses

Because SIP is largely responsible for signalling, the payload it carries does not have to be voice traffic. It can be used to handle a number of applications such as instant messaging, video conferencing and even network games.

One of SIP‘s advantages is its unique ability to return different media types within a single session. For example, a customer could call a travel agent, view video clips of possible holiday destinations, complete an on-Iine booking form and order currency - all within the same communication session.

It is important to remember that SIP does NOT transport any data. It is a signalling protocol responsible for maintaining the connection and handling the addition of other users or services to an existing connection. It can be used to transport digitised voice information, but it does not handle multimedia data. In this application, data will be transferred via conventional TCP and may not follow the same path as a SIP packet. Where the network supports it, RTP will be used

in preference to TCP.

Examples of operation

[image]

The above illustration shows a successful attempt by user Alice to establish a session with user Bob, whose URI is bgb@biloxi.cgm. Alice’s UAC is configured to communicate with a proxy server (the outbound server) in its domain and begins by sending an INVITE message to the proxy server that indicates it desires to invite Bob’s UAS into a session (1): the server acknowledges the request (2). Although Bob’s UAS is identified by its URI, the outbound proxy server needs to account for the possibility that Bob is not currently available or that Bob has moved. Accordingly, the outbound proxy server should forward the INVITE request to the proxy server that is responsible for the domain biloxi.com. The outbound proxy thus consults a local DNS server to obtain the IP address of the biloxi.com proxy server (3), by asking for the DNS SRV resource record that contains information on the proxy server for biloxi.com.

The DNS server responds (4) With the IP address of the biloxi.com proxy server (the inbound server). Alice’s proxy server can now forward the INVITE message to the inbound proxy server (5), which acknowledges the message (6). The inbound proxy server now consults a location server to determine Bob’s location (7), and the location server responds with Bob’s location, indicating the Bob is signed in, and therefore available for SIP messages (8).

The proxy server can now send the INVITE message on to Bob (9). A ringing response is sent back from Bon to Alice (10,11,12) while the UAS at Bob is alerting the local media application (for example, telephony). When the media application accepts the call, Bob’s UAS sends back an OK response to Alice (13,14,15).

Finally, Alice’s UAC sends an acknowledgement message to Bob’s UAS to confirm the reception of the final response (16). In this example, the ACK is sent directly from Alice to Bob, bypassing the two proxies. This occurs because the endpoints have learned each other’s address from the INVITE/200 (OK) exchange, which was not known when the initial INVITE was sent. The media session has now begun, and Alice and Bob can exchange data over one or more RTP connections.

[image]

The next example (illustrated above) makes use of two message types that are not yet part of the SIP standard but are documented in RFC 2848 and are likely to be incorporated in a later revision of SIP. These message types support telephony applications. Suppose that in the preceding example, Alice was informed that Bob was not available. Alice’s UAC can then issue a SUBSCRIBE message (1), indicating that it wants to be informed when Bob is available.

This request is forwarded through the two proxies in our example to a PINT (PSTN-internet Networking) server (2,3). A PINT server acts as a gateway between an IP network from which comes a request to place a telephone call and a telephone network that executes the call by connecting to the destination telephone. In this example, we assume that the PINT server logic is collocated with the location service. It could also be the case that Bob is attached to the Internet rather than a PSTN, in which case the equivalent of PINT logic is needed to handle SUBSCRIBE requests. In this example, we assume the latter and assume that the PINT functionality is implemented in the location service. In any case, the location service authorises subscription by returning an OK message

(4), which is passed back to Alice (5,6). The location service then immediately sends a NOTIFY message with Bob’s current status of not signed in (7,8,9), which Alice’s UAC acknowledges (10,11,12).

[image]

The above illustration continues the example. Bob signs on by sending a REGISTER message to the proxy in its domain (1). The proxy updates the database at the location service to reflect registration (2). The update is confirmed by the proxy (3), which confirms the registration to Bob (4). The PINT functionality learns of Bob’s new status from the location server (here we assume that they are collocated) and sends a NOTIFY message containing Bob’s new status (5), which is forwarded to Alice (6,7). Alice’s UAC acknowledges receipt of the notification (8,9,10).

SIP Messages

As was mentioned, SIP is a text-based protocol with a syntax similar to that of HTTP. There are two different types of SIP messages, requests and responses. The format difference between the two types of messages is seen in the first line. The first line of a request has a method, defining the nature of the request and a Request URI, indicating where the request should be sent. The first line of a response has a response code. All messages include a header, consisting of a number of lines, each line beginning with a header label. A message can also contain a body such as an SDP media description.

For SIP requests, RFC 3261 defines the following methods:

REGISTER: used by a user agent to notify a SIP configuration of its current IP address and the URls for which it would like to receive calls

INVITE: used to establish a media session between user agents

ACK: confirms reliable message exchanges

CANCEL: terminates a pending request, but does not undo a completed

call

BYE: terminates a session between two users in a conference

OPTIONS: solicits information about the capabilities of the callee, but does not set up a call

For example, the header of message (1) in figure 2 might look like the following:

[image]

The first line contains the method name (INVITE), a SIP URI, and the version number of SIP that is used. The lines that follow are a list of header fields. This example contains the minimum required set.

The Via headers show the path the request has taken in the SIP configuration (source and intervening proxies), and are used to route responses back along the same path. As the INVITE message leaves, there is only one header inserted by Alice. The line contains the IP address (1226.17.91), port number (5060), and transport protocol (UDP) that Alice wants Bob to use in his response.

The Max-Forwards header limits the number of hops a request can make on the way to its destination. It consists of an integer that is decremented by one by each proxy that forwards the request. If the Max-Forwards value reaches 0 before the request reaches its destination, it is rejected with a 483 (Too many hops) error response.

The To header field contains a display name (Bob) and a SIP or SIPS URI (sip:bob@biloxi.com) toward which the request was originally directed. The From header field also contains a display name (Alice) and a SIP or SIPS URI (sip:alice@atlanta.com) that indicate the originator of the request. This header field also has a tag parameter that contains a random string (1928301774) that was added to the URI by the UAC. It is used to identify the session.

The Call-ID header field contains a globally unique identifier for this call, generated by the combination of a random string and the hostname or IP address. The combination of the To tag, From tag and Call-ID completely defines a peer-to-peer SIP relationship between Alice and Bob and is referred to as a dialogue.

The CSeq or Command Sequence header field contains an integer and a method name. The CSeq number is initialised at the start of a call (314159 in this example), incremented for each new request within a dialogue, and is a traditional sequence number. The CSeq is used to distinguish a retransmission from a new request.

The Contact header field contains a SlP URI for direct communication between user agents. Whereas the Via header field tells other elements where to send the response, the Contact header field tells other elements where to send future requests for this dialogue.

The Content-Type header field indicates the type of the message body. The Content-Length header field gives the length in octets of the message body.

The SIP response types defined in RFC 3261 are in the following categories:

Provisional (1xx): the request was received and is being processed

Success (2xx): the action was successfully received, understood and accepted

Redirection (3xx): further action needs to be taken in order to complete the request

Client Error (4xx): the request contains bad syntax or cannot be fulfilled at this server

Server Error (5xx): the server failed to fulfil an apparently valid request

Global Failure (6xx): the request cannot be fulfilled at any server

For example, the header of message (13) in figure 2 might look like the following:

[image]

The first line contains the version number of SIP that is used and the response code and name. The lines that follow are a list of header fields. The Via, To, From, Call-ID and CSeq header fields are copied from the INVITE request. (There are three Via header field values — one added by Alice’s SIP UAC, one added by the atlanta.com proxy, and one added by the biloxi.com proxy.) Bob’s SIP phone has added a tag parameter to the To header field. This tag is incorporated by both endpoints into the dialogue and is included in all future requests and responses in this call.

More information

http://www.cs.columbia.edu/~hgs/sip

http://www.sipcenter.com

http://www.sipforum.com

http://www.protocols.com

Further Reading

Where Wizards Stay Up Late

The Road Ahead

Idea Man

Being Digital

• Wireless Technology