The History Of The Technological Requirements

Published Date: 02 Nov 2017

Introduction

I have written this introductory chapter to communicate why I am passionate about wireless Internet telecommunications. I hope this chapter will motivate you to read the rest of the book, by introducing an exciting vision of the future of communications, and starting on the road to explaining technologies to realize this vision. In this chapter, I also preview the rest of the book and explain the scope of coverage. Some of the terms introduced in this chapter may not be familiar to you, but they will be explained in due course.

1.1 An Exciting Future

Between the late 1970s and early 1990s, developments and inventions in two different areas� wireless and the Internet� were going on roughly at the same time, each in its own world. Especially in the 1990s, each area grew and blossomed tremendously. Second generation (2G) digital wireless personal communications phones were acquired by significant portions of the general population, becoming almost a standard personal accessory by the end of the decade. Meanwhile, the Internet also became a part of popular culture, due in large part to the success of the World Wide Web. Today, young people in their teens and twenties are among the most comfortable users of wireless and the Internet, and the next generation may be at the forefront of future advances in these areas. Consider the observation regarding contemporary college students that �they surf the Web during class, get to know their soul mates through instant messaging and talk on cell phones while biking across the Quad. Make way for the technology natives,� and the statement, �I don�t understand how people used to live before the Internet existed� [1].

Today, the Internet and wireless are converging towards an exciting future. This can be partly explained by the recent convergence of communications and computing technologies. However, it is also indicative of the possibilities that could be obtained by merging the features of the Internet and of wireless communications. Many of these features, related to quality of service (QoS), security, mobility, and multimedia traffic support, are only recently maturing, so we are on the verge of a revolutionary leap.

For example, we consider an imaginary scenario set sometime in the future involving a family of four, with Alice and Bob the parents and Charles and Diana the children. Each member of the family has a personal communications device (we call it a �communicator� for purposes of this example) that he or she carries around everywhere. In the morning, when she wakes up, Alice finds a video message or e-mail on her communicator. It has arrived overnight through the fourth generation (4G) wireless network and it�s from her secretary, reminding her of an important meeting today. She rushes off to work, while Bob prepares breakfast with a few key presses on his communicator� he has previously asked it to remember the default services for breakfast preparation, which include requests to the toaster to toast five slices of bread, the refrigerator to prepare four glasses of milk, the coffeemaker to brew a pot of coffee, and the kitchen TV set to tune in to the morning news. These communications happen over the home wireless network based on wireless local area network (LAN) or similar technology. Meanwhile, the refrigerator finds that it is running low on milk and eggs, and places an order with a local grocery store.

While Bob, Charles, and Diana are having breakfast, a TV commercial advertises discount tickets to a concert. Charles is interested to know more and speaks to his communicator about his interest. It connects him to a Web page. After browsing a few minutes, he indicates that he wishes to speak to a salesperson about the concert, and he is connected. Charles has not specified a preference for a voice-only or a video call, but in this case, the other side has a preference for voice-only, so the call is voice-only. A secure channel for this Voice over Internet Protocol (VoIP) traffic is set up, and the salesperson�s identity is authenticated, allowing Charles to feel comfortable providing his credit card number. Diana meanwhile uses her communicator to check her buddy list of friends. Each of them has their own specifications for their current availability (e.g., available for voice calls and to receive images, but not open for video calls), which are color-coded for convenience. The friends have different levels of availability because their communicators and/or subscriptions have different sets of capabilities, and also because they can choose to limit their availability at any time. Diana has also noticed that the video quality while communicating with her richer friends is usually better than the video quality while communicating with her poorer friends. This is because her richer friends are subscribing to costlier packages that provide higher qualities of service.

Meanwhile, Alice is driving to work and mentally rehearsing the points she wishes to bring up at the meeting. She finds the most convenient fast-food store from which to pick up some breakfast, using a store locator service through her communicator. The store locator service provides the answer based on Alice�s current location. After breakfast, her friend Karen calls, wanting to initiate a video call with Alice. Alice�s car is equipped to display the video images on the windshield (a �heads-up display�), so she can drive while having the video call, but she does not want to be disturbed at the moment. She requests her communicator to schedule a time after 3 p.m. when she can talk with Karen. She does not specify a particular time, except that it must be a time slot when she does not have something else already scheduled. Her communicator checks her calendar, then communicates with Karen�s communicator to make the arrangements. Meanwhile, Alice needs to check some points in an important document. She requests her communicator to download the document and display pages nine and ten on her heads-up display. Because Alice has a high-priority business subscription, the large file transfers almost instantaneously. Also, as a fringe benefit, the light classical music playing in the background on her personalized radio service provides crystal-clear sound. All this is happening as the car is moving at highway speeds along a highway and smoothly handing off between base stations of the 4G mobile wireless system, so as far as Alice is concerned, the connection is seamless even though she is moving.

Of course, this is only a projection of how the future might look. There could always be new ideas (technologies, applications, and so forth), even some disruptive technologies, that result in changes of direction, with different emphases. However, this scenario and others like it are good guides for shaping our thinking, and they provide targets for which to aim.

?

Requirements

This vision of the future requires a number of technological breakthroughs, some of which have already been made and are available today. Underlying these requirements is the cost constraint� the features need to be available at reasonable costs, to avoid being merely technology without commercial value. Furthermore, the relationship between the new services and applications, and the supporting technological innovations, is not so much a case of our having a clear idea exactly what the services and applications are and what they need, and then just solving the technical problem. Rather, the relationship is more of a continual iterative process, as shown in Figure 1.1. For example, in the early days of the Internet, when the technology was being designed, wireless services and applications were beyond the horizon. Therefore, later on, with the coming together of wireless and Internet protocol (IP), IP needed to be enhanced to support mobility. This in turn has been fueling further developments of services and applications like location-based services that can take advantage of mobility and other recent technological advancements. Similarly, even as different services and applications have driven the need for better and varied levels and types of security, the technology has matured to the point that consumer financial transactions are increasingly being conducted over the Internet, and other new and innovative services and applications are being tried out as well.

1.2.1 Technological Requirements

The requirements to realize this future include: 1. 2. 3. 4. 5.

High level of service integration; Advanced service enabling software technologies; High-rate, reliable wireless communications; Mobility support; Supporting network infrastructure providing: � Service differentiation; � Secure communications.

Except for point 1, most of these points are explored at length in subsequent chapters in this book, so we discuss point 1 here and only briefly introduce the rest of these points in this section.

The vision of the future wireless Internet presented in Section 1.1 requires a high level of service integration. When services like voice and the Internet are highly integrated, a user can click on a link on a commercial Web page to initiate a call to customer service, or a user can pick up voice and e-mail messages from either a phone or mail program, or a user can dial the same number and have it simultaneously ring both a regular phone and the �Internet phone� on the called party�s laptop. While such integrated services may have particular names like computer-telephony integration (CTI), unified messaging, or call forking, depending on what services are being integrated, the underlying idea is the integration of services that used to be separate.

Going one step further, we can argue for greater integration of communications systems, not just integration of services. This is sometimes called convergence. In fact, different convergences are occurring, including convergence of computing and communications, convergence of wireless and the Internet, and convergence of communications systems. Here I am referring to the convergence of communications systems. Traditionally, engineers have designed different communication systems separately, each system with its own intended services that are provided in its own way to its intended users. For example, there is the telephone system that provides mainly telephony services, and there are various data network systems that provide mainly data networking services. Such systems are also called �stovepipe� solutions, and they have their own way of solving the various sub problems like networking and transport. Convergence of communications systems is about knocking down the walls between the stovepipe solutions.

Clearly, the converged systems would need to be multiservice systems, which are systems that provide multiple services, such as voice and video telephony and data communications services, instead of just one or a few services, as with the stovepipe systems. A major advantage of convergence is the greater potential for service integration. A tradeoff is that each stovepipe system may be highly optimized for the few services that it provides, whereas the converged network may not be able to be so optimized for particular services, since it needs to support multiple services. For example, as traditionally circuit-switched systems for voice telephony converge with traditionally packet-switched systems for data like the Internet, the converged network could be either circuit-switched or packet-switched. Although circuit switching is generally more optimal for voice telephony, packet switching may be chosen for the converged network. Nevertheless, research is in progress in voice over packet switching (discussed in Chapter 4), to make the best of a less optimal solution.

Moving on to the next requirement, I believe that the software and software architectures that enable services in the multiservice converged networks will need to be more sophisticated than in stovepipe systems, and flexible enough to support innovative integrated services and applications. Traditionally powerful concepts in the architecting of complex systems can be applied� concepts like abstraction and layering. Thus we are seeing middleware concepts like open systems access (OSA) emerge (discussed in Chapter 10), allowing application software to make use of features like location information without needing to understand how the information is obtained. Instead, lower-layer features are abstracted for use by higher layers.

The wireless medium is notoriously difficult to work with. Signal strength can fluctuate greatly due to fundamental problems like the multipath phenomenon, shown in Figure 1.2 (the figure illustrates multipath between a transmitter and a cell phone, and between the same transmitter and a laptop). Multipath describes when the signal from a transmitter reflects off different objects and takes multiple paths to the receiver, potentially causing destructive interference to itself. It is therefore a great challenge to provide high data rates while at the same time ensuring highly reliable communications. How high are the data rates needed? How reliable should wireless communications be? Furthermore, the duration is also an issue� sustained high data rate wireless communications consumes tremendous amounts of resources. Wireless third generation (3G) systems provide higher data rates than their 2G counterparts, but these rates are not as high as the rates provided by wireless LAN systems. A number of technologies in the pipeline may provide even higher data rates for the fourth generation (4G) systems.

Wireless provides for freedom of movement of communications devices. However, that adds challenges for the network to keep track of the location of these devices and to provide for communications even as the devices move across points of attachment to the network. Mobility management is needed. A variety of schemes for QoS differentiation (see Chapter 7) have been proposed, and work is continuing to ensure that wireless IP telecommunications systems will have the necessary service quality foundations. Security is also a serious concern, especially given the additional challenges that are faced in wireless environments (compared with wireline environments). The security aspects (see Chapter 8) of these systems are closely tied with mobility aspects, because the need to handle mobility well is one of the reasons why maintaining security is challenging in wireless Internet telecommunications.

1.2 Preview

Wireless communications and Internet-based communications have been growing rapidly in recent years. In the early years of wireless cellular systems, most of the interest in wireless was focused on circuit-switched voice communications. However, the Internet has been growing, and the volume of packet-switched data traffic along with it. Moreover, 802.11-based wireless LANs have been growing in popularity, and often are used as an extension to the wired Internet. Therefore, it is making increasing economic sense for voice and data to share a common packetswitched infrastructure, with IP-based packet switching as the natural candidate for most cases. It is important to note that both 3GPP and 3GPP2 (standards groups to design systems for 3G cellular wireless and beyond, which will be discussed later in the book) are moving towards the all-IP wireless network concept.

Since many readers may be familiar with either IP or wireless, but not necessarily both, Chapters 2 and 3 provide brief tutorials on IP and Internet concepts, and on wireless networks, respectively. The coverage in these chapters is aimed at bringing the newcomer (to either IP or wireless) up to speed quickly, while at the same time touching on issues that relate to the expositions in later chapters. If you are a newcomer to either IP or wireless, you should be able to read and understand the subsequent chapters after reading the introductions in Chapters 2 and 3. Nevertheless, the references provided in these chapters will be helpful for further study and background reading in IP and wireless technology, respectively.

The wireless Internet is expected to support a variety of applications, some new and some evolved from existing wireless or Internet applications. The communication of multimedia content will be featured in many of these applications. Thus, after introducing wireless and the Internet, the book will discuss technologies related to communicating multimedia over IP. There are a variety of recent advances in technologies for packet-switched voice, video, and other multimedia. Protocols like Real-time Transport Protocol (RTP) have been developed to handle multimedia transport, addressing issues like synchronization of different components of realtime multimedia streams. Session Initiation Protocol (SIP) is gaining popularity as a flexible protocol for session control. Multimedia transport issues will be discussed in Chapter 4, whereas session control with SIP will be discussed in Chapter 5.

The three main challenges in wireless networking, viewed from a broader perspective (i.e., not just in the context of multimedia traffic), are mobility, QoS, and security. Mobility management is explored in Chapter 6. Chapter 7 discusses QoS. Chapter 8 discusses security. Meanwhile, the Internet Protocol itself is being upgraded to incorporate various features that will be more conducive for its use in future networks, including wireless IP networks. As such, topics including mobility support and support for a very large address space are included in Internet protocol, version 6 (IPv6). The services and applications driving any network technology are the keys to success. Recent advances in thinking about and understanding such topics as services and applications and middleware will be discussed in Chapter 10.

Wireless mobile systems are evolving from 2G systems to 3G systems, which will happen in phases. As we move past the first couple of phases, 3G systems will move towards being IP-based. The IP multimedia subsystem (IMS) in the Universal Mobile Transmission System (UMTS) developed by 3GPP will be introduced in Chapter 12, after we trace the evolution of the Global System for Mobile communications (GSM) to UMTS in Chapter 11. This will serve two purposes. First, of the wireless all-IP systems being developed, IMS in UMTS is the furthest along in development. Being on the cutting edge, IMS uses the latest Internet technologies, and IMS developments are also being fed back into the Internet Engineering Task Force (IETF), shaping the development of IPv6 and protocols like SIP. Therefore, understanding this system will be of great importance and interest to the reader of this book. Second, UMTS and IMS will be used to illustrate how the various components come together in an example of an all-IP system. This should help the reader to understand better how various aspects, such as session control signaling, QoS control, and security, can be put together in a real system. Where other alternatives (i.e., different from the actual choices 3GPP made in designing IMS) are possible, these will also be discussed.

While wireless IP telecommunications is still evolving, it is imperative to look forward at possibilities for future developments. This will help to impart to the reader a feeling for the evolving nature of the field, and some of the areas and topics the reader may encounter (and perhaps be actively working in) in the future. Therefore, Chapter 13 focuses on future developments and the elusive concept of 4G.

End of chapter 1 � Introduction

?

CHAPTER 3 - Wireless Networks

The growth of wireless communications in the last two decades of the twentieth century has been astounding. Despite impressive advances in wired (wireline) communications technology in the same period (e.g., in fiber-optic technology), wireless is now a permanent and prominent feature of the telecommunications landscape. People gladly pay to be free to communicate on the go without being tied down to fixed locations.

Long before the telephone was invented, wireless communications existed� in a sense� people shouting to each other over the air between them did not use wires to communicate. However, with the invention of the telephone, electrical signals carried the sound on wires, and thus began the rise of wired telecommunications. Long distance telecommunications was born. While telephony with wirelines greatly increased the range over which telecommunications was possible, it tied users to fixed locations (where the phones were located), and mobility was traded off for range. Mobile telephone systems, when they appeared, regained some of the mobility lost by wireline telephony, without sacrificing range in the end-to-end sense. 1 Since only the segment from mobile phone to base station is wireless and then it plugs into the global phone network, the corresponding party could be very far away, such as on the other side of the globe.

However, wireless links are generally of lower quality and less reliable than wired links. Why? Fundamentally, the wireless communications medium is more difficult than wired communications media. As the signals propagate through the air, they are neither guided by the structure of the wires nor shielded from interference. As a result, the signal energy dissipates rapidly, and interference from other signals can be challenging. In addition, the signals are scattered and attenuated by various objects in the environment like trees and buildings. The same signal arrives at the receiver after traversing different paths, a phenomenon known as multipath that we introduced in Chapter 1. We can only scratch the surface here; for many more details on the wireless communications medium, a classic but still useful reference is Jakes.

Thus, in the telephony world, mobile phone users trade off quality for the convenience of mobility. A similar tradeoff is seen in the data communications world.

Ethernet (IEEE 802.3), a wired LAN technology, provides reliable, high data-rate communications. Wireless LAN (WLAN) technology like �wi-fi� (a popular name for WLANs based on IEEE 802.11, a specification in the same family as 802.3) provides mobility to users, but the tradeoff is less quality and less reliability. A general principle emerges: Wireless communications is about tradeoffs first and playing catch-up with wired communications capabilities second. While wireless links continue to improve (with higher data rates and lower error rates), one may observe that they seem to be always a step behind their wired counterparts. For example, by the time WLAN technology has started moving up from 2 Mbps to 11 Mbps, 100-Mbps Ethernet is gaining momentum to replace 10-Mbps Ethernet technology.

?

3.1 Short History

The discovery of Maxwell�s equations and other ground-breaking work on electromagnetism in the nineteenth century, set the foundations for communications using electromagnetic waves. When Hertz discovered in 1886 that electromagnetic waves could propagate not just over wires but also through the air without wires, wireless communications became possible. Hertz also first demonstrated the transmission and reception of electromagnetic waves through the air. The earliest wireless communication systems were the first radio telegraphs demonstrated around the turn of the century. By the early twentieth century, voice (not just telegraphic signals) could be carried by radio. Public radio started taking off [using amplitude modulation (AM)], followed by TV. However, both of these applications of wireless communications are broadcast, one-to-many applications with a large, high-power transmitter that transmits signals to be received by thousands of receivers. Two-way person-to-person long-distance wireless communications were not widely used. Although Al Gross invented the walkie-talkie in 1938, the telephone companies showed little interest in combining wireless and telephony even in the 1950s.

A major development was the creation of the cellular concept in the late 1970s, followed by the deployment of first generation (1G) cellular systems making use of the concept. Previous to the development of cellular, two-way wireless communications systems were used in cities, but these would typically consist of one massive base station that could only support a few expensive user terminals. Multiple simultaneous communications did happen, but they used different frequency channels. Frequency channels are frequency bands used for wireless communications, and separated far apart enough from other frequency channels so that the interference between them is limited. Typically, in a scheme known as frequency division duplex (FDD), a pair of frequency channels is used together for two-way communications, one for base station to terminal (downlink or forward link) and one for terminal to base station (uplink or reverse link). Whenever a pair of frequency channels is in use for communications between a terminal and a base station, that pair of channels cannot be used simultaneously by another terminal throughout the entire city. With the cellular concept, many base stations are used, but the coverage area of each base station is limited. This allows the same pairs of frequency channels used in a cell to be reused in other cells. If all the cells using the same frequency pairs are sufficiently separated from one another, they interfere only minimally with one another. This concept allows frequency reuse, tremendously increasing the capacity of the system (the number of users). Figure 3.1 illustrates the cellular concept. On the left is the case where cells are not used. On the right, cells are used, with a frequency reuse factor of three for illustration purposes (i.e., three sets of frequencies are used, for interference reduction; other frequency reuse factors are also possible). The frequency sets are labeled numerically beneath the base station icon in each cell in the figure. Notice that adjacent cells use different sets of frequencies. Also, note that the real wireless propagation environment is not so neat� the hexagons representing coverage areas in this figure are merely a convenient representation. In reality, there would be difficulties placing base stations in such a regular set of positions, and coverage would be of varying qualities in a cell, with poorly covered locations interspersed throughout.

While the cellular concept is a breakthrough in system capacity, it introduces a new challenge� handoff. Since users are moving around, and the coverage area of each base station is limited, users inevitably need to switch between base stations, a process known as handoff. As can be seen in Figure 3.1, some of the phones are on the boundaries between cells. These phones would need to hand off between the cells. Handoff is a complex and interesting topic that will be covered further in Chapter 6.

The 1G wireless telecommunications systems are based on analog telephone technology. Voice is carried on analog circuits. Handoff between base stations is network controlled, with all the intelligence and decision-making in the network.

The electrical circuitry needed to support multiple frequency channels, handoffs, and other such functions was considered state of the art in its time in the late 1970s and early 1980s.

The 2G wireless telecommunications systems are based on digital telephone technology. These systems use digital voice coders (coder/decoders) and digitally coded information streams, as well as the latest digital signal processing. The latest hardware supports the increasing computational requirements. Handoff is also improved, with the mobiles assisting in the decision process. A concept known as soft handoff is used in some 2G systems. With soft handoff, instead of simply switching between base stations (a hard handoff), a mobile may be communicating with the system through multiple base stations at the same time during the soft handoff [2]. The mobile eventually decides on one of the base stations, drops the others, and completes the soft handoff. The idea behind soft handoff is to have a smoother handoff process to lessen the service disruptions caused by handoffs.

Interestingly, in going from 1G to 2G, Europe went from many different, country-local systems, to a single unified system, GSM, whereas the United States went 2 from one system, Advanced Mobile Phone System (AMPS), to two different systems, code division multiple access (CDMA) and time division multiple access (TDMA). 3 CDMA, specified in the Interim Standard (IS) 95 (IS-95), and TDMA, specified in IS-54 and more recently in IS-136. We discuss the GSM system in Section 3.2.1.

In 1998, 15 3G radio transmission technology (RTT) proposals from all over the world were received by the ITU. The grand goal was to end up with one RTT for the single global 3G mobile system, whether by selecting one of the proposals or merging a subset of them into one. Unfortunately, this goal was not achieved, due to political differences and deeply vested interests (various parties placed different degrees of importance towards backward compatibility with the existing incompatible 2G systems). Serious attempts were made to reconcile the differences. However, the discussions succeeded only in merging the proposals into five approved RTTs. Of these, the leading proposals were the wideband code division multiple access (WCDMA) and cdma2000 proposals. Both use wideband CDMA in the sense that the bandwidths (5 MHz or more) are wider than the 1.23 MHz used in the 2G CDMA system.

In the world of communications, there are two hemispheres, telephony and data communications (recall the divide between the bell-heads and net-heads as introduced in Chapter 1). As far as wireless communications is concerned, 1G and 2G systems came out of the telephony hemisphere. The response from the data communications hemisphere came in the form of WLANs, which arrived at a time when 2G systems were widely deployed and 3G was being planned. The rapid growth in popularity of WLANs was surprising and caught many people off guard. Companies that had spent billions on spectrum for 3G began getting worried as the WLAN segment of the market grew with astonishing rapidity.

Given the ubiquitous nature of wired LANs in the world of data networking, it is only natural to wonder if LANs could also be built upon wireless links. Although there had been previous attempts to design and sell WLANs, these had been proprietary solutions from individual companies. The solutions were not compatible with one another. Finally, the Institute of Electrical and Electronics Engineering (IEEE) worked to create an open standard that all the vendors could use. The result in 1999 was the 802.11 standard, which is explained in Section 3.2.2.

As the number of computing and communications devices has proliferated in recent years, so has the number of short cables between them. There may be short cables between a PC and a printer or other peripherals. Things would be much neater without lots of messy cabling between devices. One of the original main drivers of Bluetooth was as a cable-replacement technology. This type of idea expanded into the more general concept of wireless personal area network (WPAN). As the name suggests, a WPAN has a shorter range than a WLAN. Apart from Bluetooth, related alternatives include HomeRF, which uses infrared wireless communications. Bluetooth will be discussed further in Section 3.2.3.

In the past few years, a number of new wireless standards have been emerging, including wi-max (IEEE 802.16) and IEEE 802.20. These and other new wireless technologies will be discussed in Chapter 13.

?

Types of Wireless Networks There are different ways to categorize wireless systems. Wireless systems may use licensed or unlicensed spectrum. They have different coverage areas and data rates. Some wireless systems carry voice only, some data only, and some both voice and data. There are also wireless technologies that are less directly relevant to wireless Internet telecommunications (and therefore beyond the scope of this book), including cordless phones, walkie-talkies, and card scanners.

One of the great divides is between licensed and unlicensed spectrum. Usually, if a radio system wants to operate in a particular band of spectrum, the operator must obtain a license from the appropriate governing body. This governing body is typically a regulatory agency such as the Federal Communications Commission (FCC) in the United States, the Ministry of Public Management, Home Affairs, Posts and Telecommunications in Japan, or the Regulatory Authority for Telecommunications and Posts in Germany. The regulatory agency may charge fees (licensing fees) for the use of the desired band and impose certain rules on the use of the band. In some cases, two or more licenses may be granted for the same band(s), where the license owners need to cooperate with one another, within the rules of the regulatory agency, to use the band. Unlicensed operators are forbidden to use the band.

However, a small number of selected frequency bands have been designated as unlicensed bands and follow a different usage model. Any device can transmit and receive wireless signals in the unlicensed bands provided that the spectrum etiquette is followed. The operators need not have obtained licenses to use the unlicensed band. What is to prevent utter chaos and a flooding of the unlicensed bands that lead to excessively high interference and poor communications for all users? The answer is the spectrum etiquette� the rules for use of the unlicensed bands are typically much more demanding than the rules for use of the licensed bands. Examples of such rules are that devices use spread-spectrum transmission techniques (to reduce interference to other devices) and that they transmit below certain emission limits. Table 3.1 summarizes the comparison of licensed and unlicensed spectrum.

Unlicensed spectrum comprises the small number of unlicensed bands, and the rest of the available spectrum is licensed spectrum. Most wireless systems, including cellular systems, therefore require licenses. However, popular systems like IEEE 802.11-based WLANs and Bluetooth use unlicensed spectrum.

Coverage areas of wireless systems vary widely. Disregarding satellite systems (yes, they are wireless too, but they are out of scope of this book), the widest areas covered by terrestrial wireless systems are kilometers in diameter. These are the wireless wide area networks (WWAN) like GSM. WLAN, meanwhile, may have ranges on the order of hundreds of meters. For very short-range applications, on the order of meters, we enter the domain of WPAN. We now discuss one leading representative system each for WWAN, WLAN, and WPAN.

?

3.2.1 Wireless Wide Area Network (WWAN): GSM

The mostly widely deployed and used wireless telephony system in the world today is GSM. GSM provides a full range of services including teleservices speech, fax, and short message service (SMS). A tele-service specifies not only the data communications between, but also the terminals. In a layered communications model, this typically means the higher layers services are specified to a substantial degree. This is in contrast to lower-level bearer services where only the transport of data between two terminal-modem interfaces is specified. GSM also provides bearer services such as 13-Kbps bearer for voice traffic (which may be used by the speech teleservice) and low-rate data traffic. Additionally, since GSM Phase 2 (GSM was introduced in two major phases, as will be explained in Chapter 11), a large number of supplementary services are also available. These are basic telephony features like call forwarding and call waiting, enhanced in one or more ways for mobile telephony (e.g., call forwarding on busy is different from call forwarding on unavailable), as well as new features like advice of charging, and barring of incoming or outgoing calls according to specific criteria. The GSM lower layers use a TDMA-based approach for multiple accesses, with Gaussian minimum shift keying (GMSK) modulation.

In this section, we will discuss the following:

� The basic GSM network architecture with basic concepts and terminology;

� Location management, including registration and paging;

� Call setup, and home and roaming cases.

The GSM network architecture was originally very circuit oriented. With the introduction of GPRS and increasing use of 3G, the network is moving towards a more packet-oriented design for data and telephony. Although this book focuses on wireless Internet telecommunications, which is packet switched using IP; we begin with the original basic circuit-switched architecture of the GSM network, for the following reasons:

� Many of the design issues are common to wireless networks in general, rather than just wireless packet switched or wireless circuit switched networks. We will see certain trends emerge that are related to fundamental challenges in wireless networks in general and in handling issues including mobility and security.

� In many cases, solutions for wireless Internet telecommunications are analogous to, or are modified versions of, corresponding wireless circuit-switched telecommunications solutions. Understanding the original GSM design will help the reader appreciate the solutions in the wireless Internet case.

The basic GSM network architecture is shown in Figure 3.2. The user handsets are known as mobile stations (MS), and they communicate with the network over the GSM air interface (a set of protocols for communication over GSM wireless links, as will be explained shortly). The network in Figure 3.2 (except for the PSTN cloud on the right) can be thought of representing a public land mobile network (PLMN). Typically, a PLMN is operated by a single operator and restricted to a geographical region like a country. There may be more than one PLMN in a large country, but a subscriber to one operator�s service usually would not be able to obtain service from the other PLMNs in the area, unless there is some prior arrangement between the operators. In this case, the service would be handled as a case of roaming, which will be discussed shortly.

The GSM standards committees made a wise choice to separate the subscriber from the terminal (or equipment) in the MS. Rather than have each subscriber identified with a mobile terminal, the subscriber is identified with a subscriber identity module (SIM), a card that can be plugged into a terminal. Together, a SIM and a terminal make an MS. Each subscriber has a unique international mobile subscriber identity (IMSI) associated with his SIM, whereas each terminal has a unique

international mobile equipment identifier (IMEI). The separation of subscriber from terminal allows one subscriber to use multiple phones and terminals, including perhaps borrowed phones while visiting foreign countries, as long as he brings his SIM along with him and transfers it between phones.

The entry points into the network are known as base transceiver systems (BTS) or, more informally, base stations (BS). The BSs are specialized radio modems and not much more. Rather than putting more intelligence and control functions in the BSs, the GSM designers added another network element, the base station controller (BSC). A BSC with the BSs it controls is referred to as a base subsystem (BSS). On the other side, several BSCs may connect to a mobile switching center (MSC). The MSC is a big machine, handling at any given time the MSs that are using the BSs and BSCs connected to the MSC. It can be thought of as a regular digital switch with added functionality to handle mobile subscribers. For example, the MSC is concerned with radio resource management for the changing set of MSs it handles (unlike a class 4 wireline telephony switch, where the set of subscribers is fixed except for added or canceled subscriptions). Like other digital switches, the MSC is a part of the global signaling system 7 (SS7) network, which uses ISDN protocols.

The added signaling over the SS7 network to handle mobile subscribers is known as the mobile application part (MAP). MAP signaling involves another two elements, the home location register (HLR) and visitor location register (VLR). The HLR is present in every PLMN, and is the database where subscriber information is stored. Even though some operators operate PLMNs in multiple regions or countries, each subscriber signs up for service in one region, so one of the PLMNs is the home PLMN. Each of the other PLMNs becomes the visited PLMN, with reference to a particular subscriber, when that subscriber tries to obtain service from it. The capability of a subscriber to obtain service from a visited PLMN is known as roaming. The VLR is used to store information on roaming users. The VLR is often integrated with the MSC (we refer to it as an MSC/VLR in such a case). The remaining two network elements in Figure 3.2 are the authentication center (AuC) and equipment identity register (EIR). The AuC is involved in authenticating MSs, a process that will be explained in Chapter 8. The EIR is used to support a theft protection mechanism. It maintains some lists of IMEIs, which can be used to raise a red flag when a terminal reported stolen is used.

3.2.1.1 Location Management

We now look at the concepts of location management, which is part of the solution to roaming (it also supports other goals such as power savings). Location management is a broad concept that includes registration, including storing parts of the MS location information in various databases (HLRs and VLRs) in the network, and the location update and paging procedures. All BSs broadcast their unique cell ID and location area ID (location areas will be introduced in a couple of paragraphs) to assist MSs in location management.

Registration is a procedure in which an MS signals with the GSM network indicating where it is located and that it is �on� and wishes to �attach� to the network. The technical term for registration in GSM is IMSI attach, and for deregistration is IMSI detach. Recall that the IMSI uniquely identifies a subscriber. If a subscriber is not registered (IMSI detached), the network knows not to bother trying to set up a call to the subscriber. This is a valuable saving of network resources (including avoiding having to page the subscriber, a subject to be discussed shortly). If a subscriber is registered (IMSI attached), the network knows it is worth trying to set up an incoming call to the subscriber. Furthermore, the network acquires knowledge of the location of the user in the process of registration. If a roaming user tries to register in a foreign network, the HLR of the user will be queried and the user authenticated, and then the VLR in the foreign network will obtain part of the subscriber profile and other subscriber-related information (including security related information, as will be elaborated on in Chapter 8) from the HLR of the user. The HLR of the user will also be set up to point to the MSC in the foreign network.

What about location update and paging? The registration described in the preceding paragraph satisfies the necessary location management, as long as the MS does not move after it registers. However, we cannot make that assumption. The MS must be allowed to move, and the network must be updated when it does move. One possibility is to inform the network whenever the MS decides to use a different base station to access the network. When there is active communications going on, this is called handoff to a different base station and is clearly the right thing to do. When active communications does not exist and the MS is in idle state, the MS using a different base station is said to be camping. It is not immediately clear if informing the network is the right thing to do whenever the MS switches from camping on one base station to camping on another. In fact, the idea behind idle state is to save power while the MS is idle. Hence, in idle state, the MS only updates it location with the network when it crosses boundaries between location areas. Location areas are groups of multiple base stations, and the MS knows when it has changed location areas by listening to the base station broadcasts. The use of location areas saves power by not requiring the MS to send location updates whenever it crosses cell boundaries. However, the network now only knows the location of the MS to the precision of a location area. Hence, when an incoming call arrives, it needs to search for the MS within the entire location area. This procedure is called paging. There is an interesting tradeoff between resource utilization for location updates and for paging. The larger the location areas, the more power savings for location updates but the more resource utilization for paging, while the smaller the location areas, the less power savings for location updates but the less resource utilization for paging.

3.2.1.2 Call Setup Signaling To illustrate how roaming is supported in call setup signaling, we first explain how a subscriber would make or receive calls when at home (in the home PLMN). We can then point out the differences in the roaming case. We use call flow diagrams to show the signaling. These diagrams show the relevant network elements at the top, followed by messages sent between them. The vertical lines are to indicate the source and destination of the messages. If the tail of an arrow is touching a vertical line beneath a network element, that element is the source of that message, whereas if the head of an arrow is touching a vertical line beneath a network element, that element is the destination of that message. The sequence of the messages is starting from the top, so the horizontal arrows are arranged in time sequence, with the later messages further down the diagram than earlier messages. In all cases, we designate the calling party as the caller and the called party as the callee.

When a subscriber in her home network dials a callee, the MSC in the caller�s home PLMN analyzes the digits and routes it to the appropriate destination, as shown in Figure 3.3 (you may want to look ahead to the Appendix 5A.1 in Chapter 5 for a brief introduction to signaling in the phone network). The destination may be in the PSTN, another PLMN, or the same PLMN. The signaling between MSC and PSTN switches is very similar to that between digital switches in the wireline PSTN. Except for the two MAP messages (messages 2 and 3), the signaling is basically the ISDN signaling used in the modern digital PSTN (more details in Chapter 5). If the callee is also a mobile subscriber, then the destination side of the signaling will be as described next.

When another party (whether wireline or mobile) dials a mobile subscriber, the call setup signaling gets routed to the home PLMN of the subscriber. It must be routed to the home PLMN even if the subscriber is roaming, because other networks do not know whether the subscriber is at home or roaming. In a PLMN with multiple MSCs, one of them may be configured as the gateway MSC. The gateway MSC in a PLMN is the entry point into the PLMN for calls destined to one of its subscribers. The gateway MSC has the responsibility to query the HLR for the subscriber profile and location information. If the subscriber is in her home PLMN, the gateway MSC just has to route the call signaling to the appropriate MSC serving the subscriber.

What about when the MS is roaming? When the MS tries to set up a call, the network that is roamed to will assist with authentication and call setup, based on the protocols and information exchanged earlier during IMSI attach. Otherwise, there is little difference from the case when the MS is at home. However, the situation becomes more interesting when the MS is receiving a call while roaming. In this case, illustrated in Figure 3.4, 4 the call setup reaches the gateway MSC, as when the MS is at home. The gateway MSC will query the HLR, and will find that the MS is roaming, as well as in which network it is roaming. These pieces of information are stored as part of registration and updated by location updates; otherwise, if the MS has not registered in the foreign network, the network will be unable to locate the MS. Call setup continues with the next call leg between the gateway MSC and the appropriate MSC in the foreign network. That MSC will route the call towards the appropriate BSC and eventually to the MS. Note that this kind of call setup could result in the voice circuits being much longer than necessary. For example, if the caller and callee are both in one PLMN, but the callee is merely roaming in that PLMN and if the callee�s home PLMN is the other side of the globe, the call will still be routed back to the callee�s home PLMN, causing long and unnecessary call legs.

Wong, Daniel (Author). Wireless Internet Telecommunications.

Norwood, MA, USA: Artech House, 2004. p 46.

http://site.ebrary.com/lib/auemirates/Doc?id=10081946&ppg=60

Summary

We switch gears coming to this chapter from the last chapter, going from the Internet to wireless technologies. The wireless communications medium is difficult, so wireless links are generally less reliable and of lower quality than wired links, but wireless provides features, such as mobility, that are very attractive. A brief history of wireless is provided, where we trace the development of cellular mobile systems in particular, including the cellular concept, frequency reuse, and handoffs. We also survey the characteristics of 1G, 2G, and 3G systems. However, while cellular mobile systems are a product of the telephony world, other wireless systems (such as WLANs) are a product of the data-networking world. We introduce these systems as well, and consider three examples of wireless systems: GSM, 802.11 WLAN, and Bluetooth. These are examples, respectively, of a wide-area network system, a local-area network system, and a personal-area network system.