Introduction to Mobile Computing

2
and Wireless Networking

A few important breakthroughs achieved over the last few decades

have revolutionized the way people use computers. First, about two
decades ago the advancements made in the field of miniaturization of
electronic circuits made it possible to pack powerful processing units and
significant memory into portable laptops. More recently, the computing
elements have become more powerful and at the same time shrunk to fit
into palmtops that people can carry effortlessly. In another development,
high speed data communication facility has been made available to
portable computing platforms, largely through advancements in the areas
of computer communication and wireless networking technologies. These
two developments have contributed to the formation of the discipline of
mobile computing. This has made it possible for the users to take advantage
of a host of innovative services available in these platforms and use them
ubiquitously. As a result, people from all walks of life have started to carry
mobile handsets with them wherever they go and are able to perform
meaningful computations with them.

2.1 What Is Mobile Computing?

Mobile computing (sometimes called ubiquitous computing and also at times

called nomadic computing) is widely described as the ability to compute
remotely while on the move. This is a new and fast emerging discipline
that has made it possible for people to access information from anywhere
and at anytime. We can also view mobile computing as encompassing
two separate and distinct concepts: mobility and computing. Computing
denotes the capability to automatically carry out certain processing related
to service invocations on a remote computer. Mobility, on the other hand,
provides the capability to change location while communicating to invoke
24
 Introduction to Mobile Computing and Wireless Networking 25

computing services at some remote computers. The main advantage of this

type of mobile computing is the tremendous flexibility it provides to the
users. The user need not be tethered to the chair in front of his desktop, but
can move locally or even to far away places and at the same time achieve
what used to be performed while sitting in front of a desktop.

2.2 Mobile Computing vs. Wireless Networking

We must distinguish between mobile computing and wireless networking.

These two terms are not synonymous. While mobile computing essentially
denotes accessing information and remote computational services while
on the move, wireless networking provides the basic communication
infrastructure necessary to make this possible. Thus, we can say that
mobile computing is based on wireless networking and helps one to invoke
computing services on remote servers while on the move: be it be office,
home, conference, hotel, and so on.
It should be clear that wireless networking is an important ingredient
of mobile computing, but forms only one of the necessary ingredients
of mobile computing. Mobile computing also requires the applications
themselves—their design and development, and the hardware at the client
and server sides. In fact, we can say that mobile computing subsumes the
area of wireless networking. Consequently, to be able to understand the
subtle issues associated with mobile computing, in addition to studying
the different aspects of mobile computing applications, their design
and development, we need to have a good knowledge of the basics of
wireless communications technologies. Wireless networking is increasingly
replacing traditional networks because of the low setup time and low initial
investment required to set up the wireless network. As we discuss later in
this chapter, wireless networks appear in various forms such as WLANs
(Wireless LANs), mobile cellular networks, personal area networks (PANs),
and ad hoc networks, etc.
Wireless networks can be classified into two basic types. One is an
extension of wired networks. It uses fixed infrastructures such as base
stations to provide essentially single hop wireless communication with
a wired network as illustrated in Fig. 2.1 or a two-hop wireless cellular
communication with another mobile as explained earlier in Fig. 1.4.
The other type of wireless network is an ad hoc network. An ad hoc
network does not use any fixed infrastructure and is based on multi-hop
wireless communication as shown in Fig. 2.2.
One popular example of a fixed infrastructure wireless network is a
Wireless LAN (WLAN) that implements the IEEE 802.11 protocol. Observe
from Fig. 2.1 that only the last hop is through the wireless medium. An
access point (AP) provides the last hop connectivity of the mobile nodes
to a wired network. All communication goes through APs which perform
26 Fundamentals of Mobile Computing

Figure 2.1 Wireless network based on fixed infrastructures.

Figure 2.2 Wireless network having no fixed infrastructures.

bridging† between the wireless and the wired mediums. A station must be
recognized by an AP to be able to connect to the network. The AP may
require authentication and this in turn is used as the basic means to keep out
the unauthorized users. In an infrastructureless network, the communication
between hosts occurs directly or via a few intermediate nodes that form
the hops. For example, station A in Fig. 2.2 can communicate with station
C using either the hops A–B, B–C or A–D, D–C.
A recent development in this context, is wireless networking of various
types of devices using the Bluetooth technology. As already discussed in
Chapter 1, the Bluetooth technology can also be used to establish direct
wireless connection of cell phones with devices such as printers, cameras,
scanners, laptop and desk computers. Bluetooth is gradually replacing
cables and infrared as the dominant way of exchanging information
between devices. One of the objectives of the Bluetooth technology is to
enable users to easily connect to a wide range of personal computing and
telecommunication devices, without the need to buy, carry, or lay out
cables. In fact, the Bluetooth technology enables setting up of personal
area networks (PANs) known as piconets and ad hoc networks known
as scatternets. It provides opportunities for rapid deployment of ad hoc
connections, and the possibility of automatic, transparent connections
between devices. It promises to eliminate the need to purchase additional
or proprietary cabling and configuration exercises needed to connect the
individual devices.
†
A network bridge connects multiple network segments at the data link layer
(Layer 2) of the OSI reference model.
 Introduction to Mobile Computing and Wireless Networking 27

An ad hoc network is also known as a Mobile Ad hoc Network(MANET).

It is a collection of mobile nodes that form a network on the fly without
requiring the support of any fixed infrastructure. Wireless sensor networks
are a special type of wireless ad hoc networks. Mobile ad hoc networks
and sensor networks are discussed in more detail in Chapters 7 and 8
respectively.

2.3 Mobile Computing Applications

Mobile computing technology makes it possible for people to send or extract

information while on the move. For example, a stock broker travelling in a
car may wish to issue stock transaction orders from a mobile phone or to
receive share price quotations. As can be guessed, ease of deployment and
scalability are two important positive points in favour of data transmissions
over the wireless medium. But, it is not without some shortcomings. When
data is being transmitted on air, all the wireless devices present in the
transmission range can receive the data. This, therefore, opens up very
difficult security issues that must be overcome to ensure privacy of data.

2.4 Characteristics of Mobile Computing

A computing environment is said to be “mobile”, when either the sender

or the receiver of information can be on the move while transmitting
or receiving information. The following are some of the important
characteristics of a mobile computing environment.
Ubiquity: The dictionary meaning of ubiquity is present everywhere. In
the context of mobile computing, ubiquity means the ability of a user to
perform computations from anywhere and at anytime. For example, a
business executive can receive business notifications and issue business
transactions as long he is in the wireless coverage area.
Location awareness: A hand-held device equipped with global positioning
system (GPS) can transparently provide information about the current
location of a user to a tracking station. Many applications, ranging from
strategic to personalized services, require or get value additions by location-
based services For example, a person travelling by road in a car, may need
to find out a car maintenance service that may be available nearby. He can
easily locate such a service through mobile computing where an application
may show the nearby maintenance shop. A few other example applications
include traffic control, fleet management and emergency services. In a
traffic control application, the density of traffic along various roads can
be dynamically monitored, and traffic can be directed appropriately to
28 Fundamentals of Mobile Computing

reduce congestions. In a fleet management application, the manager of a

transport company can have up-to-date information regarding the position
of its fleet of vehicles, thus enabling him to plan accurately and provide
accurate information to customers regarding the state of their consignments.
Location awareness can also make emergency services more effective by
automatically directing the emergency service vehicles to the site of the call.

Adaptation: Adaptation in the context of mobile computing implies

the ability of a system to adjust to bandwidth fluctuation without
inconveniencing the user. In a mobile computing environment, adaptation
is crucial because of intermittent disconnections and bandwidth fluctuations
that can arise due to a number of factors such as handoff, obstacles,
environmental noise, etc.

Broadcast: Due to the broadcast nature of the underlying communication

network of a mobile computing environment, efficient delivery of data can
be made simultaneously to hundreds of mobile users. For example, all
users at a specific location, such as those near a railway station, may be
sent advertising information by a taxi service operator.

Personalization: Services in a mobile environment can be easily

personalized according to a user’s profile. This is required to let the users
easily avail information with their hand-held devices. For example, a
mobile user may need only a certain type of information from specific
sources. This can be easily done through personalization.

2.5 Structure of Mobile Computing Application

A mobile computing application is usually structured in terms of the

functionalities implemented. The simple three-tier structure of a mobile
computing application is depicted in Fig. 2.3. Figure 2.4 shows a specific
scenario of the types of functionalities provided by each tier. As shown in
these figures, the three tiers are named presentation tier, application tier
and data tier.
We now briefly explain the roles of the three tiers of a mobile computing
application.

Presentation (Tier-1)

Application (Tier-2)

Data (Tier-3)

Figure 2.3 The three tier structure of a mobile computing application.

 Introduction to Mobile Computing and Wireless Networking 29

Figure 2.4 Functionalities provided by each tier structure of a mobile computing

application.

Presentation tier
The topmost level of a mobile computing application concerns the user
interface. A good user interface facilitates the users to issue requests and
to present the results to the them meaningfully. Obviously, the programs
at this layer run on the client’s computer. This layer usually includes web
browsers and customized client programs for dissemination of information
and for collection of data from the user.

Application tier
This layer has the vital responsibility of making logical decisions and
performing calculations. It also moves and processes data between the
presentation and data layers. We can consider the middle tier to be
like an “engine” of an automobile. It performs the processing of user
input, obtaining information and then making decisions. This layer is
implemented using technology like Java, .NET services, cold fusion, etc.
The implementation of this layer and the functionality provided by this
layer should be database independent. This layer of functionalities is
usually implemented on a fixed server.
30 Fundamentals of Mobile Computing

Data tier
The data tier is responsible for providing the basic facilities of data
storage, access, and manipulation. Often this layer contains a database.
The information is stored and retrieved from this database. But, when only
small amounts of data need to be stored, a file system can be used. This
layer is also implemented on a fixed server.

2.6 Cellular Mobile Communication

In a cellular mobile system, the area of coverage is split into cells as shown
in Fig. 2.5. Even though the cells have been shown to be non-overlapping
hexagons for simplicity, but in reality cell shapes are irregular and do
overlap to some extent. A base station (BS) is located at the centre of each
cell. The BS in a cell receives communications from all mobile handsets
in the cell and forwards the data to the appropriate handset. Thus, a
base station keeps track of the calls of all handsets in its cell. When a
mobile handset while still continuing a call, moves to another cell, the BS
“hands-off” the call to the BS in the new cell. When a cell covers a crowded
area having too many users, then the users can experience frequent call
drops. To overcome this problem, such cells are split into smaller cells.

Figure 2.5 Cell structure in a cellular mobile communication system.

Initially the focus of cellular mobile communication was voice

communication. But today cellular phones provide many services based on
data communication too. These include electronic mail, Internet access and
running a variety of mobile applications. The term mobile communication
has a much wider connotation than that of cellular communication, and
includes wireless LANs and ad hoc networks. However, due to the
overwhelming popularity of mobile phones, cellular communication and
mobile communication are at times used interchangeably.
 Introduction to Mobile Computing and Wireless Networking 31

2.6.1 Generations of Cellular Communication Technologies

Mobile communication technology has advanced at a very rapid pace

over the last five decades. The gradual technology improvements over the
last four decades can be roughly demarcated into four generations. Each
generation essentially provides higher data rate and additional capabilities,
as shown schematically in Fig. 2.6. This figure does not show the data rates
of technologies before GSM, since these were analog techniques that did
not support the data communications facility. The fourth generation (4G) of
technology provides a substantial order of magnitude improvements in data
speeds, but is not yet widely implemented. The important characteristics of
the various generations of cellular mobile systems have been summarized in
Table 2.1. As can be seen from the table, each passing generation of mobile
cellular system brought about significant advancements to the technology,
causing the quality of the services to improve and the number of service
offerings to increase, and at the same time the cost to the customer to drop
drastically. We briefly discuss these different generations of mobile cellular
communication systems in the following.

Figure 2.6 Summary of mobile technology advancements.

First generation
The first generation (1G) cellular system was designed in the late 1960s,
but was commercially deployed in the early 1980s. The first commercial 1G
system in the United States was known as Advanced Mobile Phone System
(AMPS). It became operational in 1982 and supported only voice calls.
This was a completely analog system. In an analog system, analog signals
are transmitted by modulating them on a higher frequency carrier signal,
without first converting the signal into digital form through quantization.
In a completely analog system, it is difficult to support SMS and other data
services. Also, the signals from different users cannot be intermixed on the
same channel, and have to be transmitted in clearly separated channels.
32 Fundamentals of Mobile Computing

TABLE 2.1 A Summary of the Important Characteristics of the Different

Generations of Cellular Wireless Communication Systems

Mobile
Features Standards Data speed
gen.
1G Analog transmissions, pri- NMT, AMPS, 600–1200 bps
marily supported voice TACS
communications.
2G Digital transmissions, impro- GSM 9.6 kbps
ved performance by letting
multiple users share a single
channel.
2.5G Enhanced multimedia and GPRS 28 kbps or
streaming video, web browsing. higher
3G Enhanced multimedia and UMTS, HSPDA, 384 kbps or
streaming video capabilities. EDGE, W-CDMA higher

In the 1G system, the available frequency spectrum was split into

a number of sub-bands (or channels), each of which was used by a
different caller. These systems typically allocated one 25 MHz frequency
band for the signals to be sent from the base station to the handset
(incoming signal), and a second different 25 MHz band for the signal
transmitted from the handset to the base station (outgoing signal).
Figure 2.7 shows a frequency band split into five sub-bands (channels).
Though for simplicity, we have shown the different channels to be adjacent
to each other, each channel was separated from the adjacent channels by a
spacing of about 30 kHz. This was called a guard band. The use of guard
bands was one of the causes of inefficient spectrum usage and resulted in
the reduced number of simultaneous calls that could be supported. This
problem was overcome in the subsequent generations of technologies.

Sub-band 1
Sub-band 2
Band Sub-band 3
Sub-band 4
Sub-band 5
Time
Figure 2.7 1G frequency band split into five sub-bands.

Different 1G standards were used in different countries:

• AMPS (Advanced Mobile Phone System) in the USA
• NMT 450 (Nordic Mobile Telephone) in various European countries
• TACS (Total Access Communications System) in the UK
 Introduction to Mobile Computing and Wireless Networking 33

The 1G systems were of multiple access type, since once a caller hanged
up, another caller could use the same frequency. For this reason, the 1G
technology was also called Frequency Division Multiple Access (FDMA).
It was possible to reuse the same frequencies in the non-adjacent cells,
because the transmitter power output was restricted. For example, the cells
shown shaded in Fig. 2.8 could use the same set of frequencies.

Figure 2.8 Cell structure and frequency allocation of 1G systems.

When a caller crossed a cell boundary, the channel being used might
not be made available in the new cell as it might already be in use in some
other cell. During handoff, a different channel was possible to be allocated
in this case, otherwise the call got dropped if none were available. Beside
the number of callers that could be accommodated being low, the voice
quality was poor due to analog transmission. Also, it provided no security
at all, since any one could hear a call by tuning into a channel.

Second generation
As already pointed out, the 1G technology had many disadvantages. The
major drawback was the small number of simultaneous calls that could
be made and the high risk of call drops during handoffs. Calls in 1G
were expensive because of the inherent inefficient usage of the bandwidth
spectrum and hence very few could afford to use a cell phone. Further, the
1G networks were not capable of providing several useful services such as
caller identity and SMS. The disadvantages of 1G systems were overcome
by the second generation (2G) cellular systems.
The 2G systems encoded voice and other information digitally before
transmitting them. Digital transmission has many advantages over analog
transmissions. These include noise immunity and better bandwidth utiliza-
tions. The 2G system offered significant advancements in the evolution of
the mobile cellular technologies. Hence the 2G technology rapidly replaced
the 1G technology because of the drastic reductions in the cost of phone
calls and availability of a wider range of services coupled with substantial
improvements in the quality of services. Also, SMSs became possible.
34 Fundamentals of Mobile Computing

However, we must remember that the 2G technology is in many

respects an extension of the 1G system and many of the principles involved
in a 1G system also apply to 2G. For example, they both use the same cell
structure. However, there are many differences. For example, they use
different signal modulation techniques. 2G uses CDMA (Code Division
Multiple Access) and TDMA (Time Division Multiple Access) as channel
access technology, while 1G used FDMA.
The 2G mobile system deployment started in the 1990s, and two
competing standards existed. In North America, the IS-95 standard was
adopted which used Code Division Multiple Access (CDMA) and could
multiplex up to 64 calls per channel in the 800 MHz band. In Europe and
elsewhere, operators adopted the Global System for Mobile Communication
(GSM) standard, which used Time Division Multiple Access (TDMA) to
multiplex up to 8 calls per channel in the 900 and 1800 MHz bands.
The first commercial deployment of Global System for Mobile
Communication (GSM) was done in 1992. It supported higher voice quality
and provided data services such as SMS and e-mail. We will discuss
the GSM system in more detail, later in this chapter. In 1993, another
2G system, known as CDMAone, was standardized and commercially
deployed in South Korea and Hong Kong in 1995, followed by the United
States of America in 1996.

2.5 Generation
General Packet Radio Service (GPRS) is an extension of GSM and is
considered to be the 2.5 generation technology. As indicated by the name,
it is based on packet switching compared to circuit switching used in 2G.
This was a significant improvement over 2G and helped to reduce call
costs dramatically. Another important advantage of GPRS is that it allows
users to remain connected to the Internet without incurring additional
charge and supports multimedia capabilities including graphics and video
communications. GPRS deployments began in 2000, followed by EDGE
in 2003. EDGE enhances the capabilities of GPRS, allows faster Internet
browsing, and makes it possible to use streaming applications. Though
this technology provided faster data rates over 2G systems, it is called
2.5G because it did not offer the multi-megabit data rates which are the
characteristics of the 3G system.

Third generation
The 3G systems are often referred to as IMT-2000 (International Mobile
Telecommunications-2000) systems since this was made a global standard
by ITU. The 3G systems support much higher data transmission rates and
offer increased bandwidth, which makes them suitable for high-speed
data applications as well as for high quality traditional voice calls. The
 Introduction to Mobile Computing and Wireless Networking 35

3G systems can be considered to be purely data networks, since voice

signals are converted to digital data, this results in speech being dealt
with in much the same way as any other form of data. The 3G systems
use packet-switching technology, and provide cheaper calls while giving
better average call quality than that of the 2G systems, but they do require
a somewhat different infrastructure compared to the 2G systems. The 3G
networks made it possible for service providers to offer many innovative
applications and services such as email, instant messaging and video
telephony, multimedia gaming, live-video buffering, and location-based
services among others. The first 3G network was deployed in Japan in
2001 by DoCoMo.
UMTS (Universal Mobile Telephone System) is one of the 3G mobile
systems that was developed within the ITU’s IMT-2000 framework. UMTS
was developed mainly for the GSM networks, so that these could be
easily upgraded to UMTS networks. In UMTS, coverage is provided by a
combination of a variety of cell sizes ranging from “in building” pico cells
to global cells provided by satellites.
Even though it was expected that the UMTS specification would become
a single global standard for 3G systems, it did not turn out that way. Now
many different versions of 3G systems have come into existence and each
one evolved from some existing 2G system. The main 3G technologies that
are prevalent include UMTS and CDMA2000. European countries have
adopted UMTS, while the USA uses CDMA2000.

Fourth generation
A 4G system provides a faster data rate than that of 3G (at least 10 times
faster) and makes mobile broadband Internet access possible. The 4G
system has made possible high speed Internet access from smartphones
and laptops with USB wireless modems. A few applications that could
not be supported in earlier generations of the cellular phone systems,
have now become possible in 4G including IP telephony, gaming services,
high-definition mobile TV, video conferencing and 3D television. The 4G
technology is expected to help solve the last mile problem that prevents
the mobile users from running applications that are available on wired
networks. There are at present two competing 4G standards: Mobile
WiMAX standard and the Long Term Evolution (LTE) standard.
In the following section, we provide a brief overview of the working of
a few cellular wireless technologies that are being popularly used at present.

2.7 Global System for Mobile Communications (GSM)

GSM (Global System for Mobile Communications) is at present being used
in India. It is possibly the most successful digital mobile system to have
ever been used till now. An important characteristic of the GSM system
36 Fundamentals of Mobile Computing

is that it provides data services in addition to voice services, and yet is

compatible to 1G systems.
GSM networks operate in four different radio frequencies. Most GSM
networks operate either in the 900 MHz or in the 1800 MHz frequency
bands. Some countries in the American continent (especially the USA and
Canada) use the 850 MHz and 1900 MHz bands because the 900 MHz and
1800 MHz frequency bands are already allocated for other purposes. The
relatively rarely used 400 MHz and 450 MHz frequency bands are assigned
in some countries, notably Scandinavia where these frequencies were
previously used for the first generation systems. In the 900 MHz band,
the uplink frequency band is 890–915 MHz, and the downlink frequency
band is 935–960 MHz.

2.7.1 GSM Services

GSM provides three main categories of services. These are:
(i) Bearer services
(ii) Teleservices
(iii) Supplementary services
In the following, we elaborate these different categories of services.

Bearer services
Bearer services give the subscribers the capability to send and receive data
to/from remote computers or mobile phones. For this reason, bearer services
are also known as data services (see Box 2.1). These services also enable the
transparent transmission of data between GSM and other networks like
PSTN, ISDN, etc. at rates from 300 bps to 9600 bps. These services are
implemented on the lower-three layers of the OSI reference model. Besides
supporting SMS, e-mail, voice mailbox, and Internet access, this service
provides the users with the capability to execute remote applications. GSM
supports data transfer rates of up to 9.6 kbps.

BOX 2.1 GSM bearer services

The GSM data services are named bearer services. Consider the following
example: Suppose a customer requires to send a data file such as a picture to
a computer at the office that is connected to a public telephone network. In
this example, the bearer service provides 9.6 kbps circuit-switched data transfer.
The handset dials the office computer telephone number and establishes a
connection with it via the modem. When the office computer modem accepts
the call, the customer’s handset begins to send data directly on the telephone
line channel at 9.6 kbps.

Bearer services permit either transparent or non-transparent, and either

synchronous or asynchronous modes of data transmission. We elaborate
these in the following.
 Introduction to Mobile Computing and Wireless Networking 37

• The transparent bearer services use the functions of the physical

layer of transmission of data leading to constant delay and
throughput if no transmission errors occur. There is a mechanism
called FEC (Forward Error Correction) to increase the quality of data
transmission.
• The non-transparent bearer services use protocols of the second and
third layers to implement error correction and flow control. They
use transparent bearer services in addition to a Radio Link Protocol
(RLP). This protocol comprises mechanisms of high level data link
control.

Teleservices
GSM provides both the voice-oriented teleservices and the non-voice
teleservices, as discussed below.

Telephony: The main goal of GSM was to provide high quality digital
voice transmission, offering the bandwidth of 3.1 kHz of analog phone
systems. Special codecs are used for voice transmission, while other codecs
are used for the transmission of analog data for communication with
traditional computer modems used in fax machines.
Emergency number: The same number is used throughout an area. This
service is free of cost and mandatorily provided by all service providers.
This connection will automatically be set up with the closest emergency
centre.
Short message services: This service offers transmission of text messages
of sizes up to 160 characters. SMS services use the signalling channels,
making possible the duplex system of the sending and receiving the SMSs
messages.
Fax: In this service, using modems fax data is transmitted as digital
data over the analog telephone network according to the ITU-T Standards
T.4 and T.30.

Supplementary services
GSM provides certain supplementary services such as user identification,
call redirection, and forwarding of ongoing calls. In addition, standard
ISDN features such as ‘close user groups’ and ‘multiparty’ communication
are available.

2.7.2 System Architecture of GSM

A GSM system consists of three main subsystems:

(i) Radio Subsystem (RSS)
38 Fundamentals of Mobile Computing

(ii) Networking and Switching Subsystem (NSS)

(iii) Operation Subsystem (OSS)
A schematic of the functional architecture of a GSM system is shown in
Fig. 2.9. The different components of this architecture are briefly explained
in the following.

Figure 2.9 Functional architecture of a GSM system.

Radio subsystem (RSS)

This subsystem comprises all the radio specific entities. That is, the mobile
stations, the base station subsystems, the base transceiver station and the
base station controller. We briefly explain the important components of
the radio subsystem in the following:

Mobile Station (MS): A mobile station (MS) or cell phone contains two
major components: the subscriber identity module (SIM) and the mobile
device. The SIM is a removable smart card. Each mobile device has a
unique identifier that is known as its IMEI (International Mobile Equipment
Identity). Apart from the telephone interface, an MS also offers other types
of interfaces to the users such as USB, Bluetooth, etc. Despite its small size,
a SIM card is a very important component of a GSM network. It contains all
the subscription information of a subscriber and holds the key information
that activates the phone after it is powered on. It contains a microcontroller
to primarily store and retrieve data from the flash storage on the SIM.
Identification information is stored in the SIM card’s protected memory
(ROM) that is not accessible or modifiable by the customer. Additional
flash memory is included in the mobile device to allow storage of other
information such as addresses, pictures, audio and video clips, and short
 Introduction to Mobile Computing and Wireless Networking 39

messages. The SIM card contains many other identifiers and tables such
as card type, serial number, a list of subscribed services, and a Personal
Identity Number (PIN).
Base Station Subsystem (BSS): A GSM network comprises many BSSs.
Each BSS consists of a Base Station Controller (BSC) and several Base
Transceiver Stations (BTSs). We will explain these components subsequently.
A BSS performs all functions necessary to maintain radio connections to
an MS, as well as does coding/decoding of voice.

Base Transceiver Station (BTS): A BTS comprises all radio equipment

such as antenna, signal processors and amplifiers that are necessary for
radio transmission. It encodes the received signal, modulates it on a carrier
wave, and feeds the RF signals to the antenna. It communicates with both
the mobile station and the BSC.
Base Station Controller (BSC): A BSC manages the radio resource of the
BTSs in the sense that it assigns frequency and time slots for all MSs in
the area. It also manages the handoff from one BTS to another within the
BSS. The BSC also multiplexes the radio channels onto the fixed network
connection to the Mobile Switching Centre (MSC).

Network and switching subsystem (NSS)

This subsystem forms the heart of the GSM system. It connects the wireless
networks to the standard public networks and carries out usage-based
charging, accounting, and also handles roaming. NSS consists of a switching
centre and several databases as described below.

Mobile Switching Center (MSC): An MSC can be considered to form

the heart of a GSM network. An MSC sets up connections to other MSCs
and to other networks such as Public Data Network (PDN). An MSC is
responsible for the connection setup, connection release, and call handoff to
other MSCs. A Gateway MSC (GMSC) is responsible for gateway functions,
while a customer roams to other networks. It also performs certain other
supplementary services such as call forwarding, multiparty calls, etc.
Home Location Registers (HLRs): A HLR stores in a database important
information that is specific to each subscriber. The information contains
subscriber’s IMSI, pre/post paid, user’s current location, etc.
Visitor Location Register (VLR): It is essentially a temporary database that
is updated whenever a new MS enters its area by roaming. The information
is obtained from the corresponding HLR database. The function of the VLR
is to reduce the number of queries to the HLR and make the user feel as
if he were in his home network.
40 Fundamentals of Mobile Computing

Operation subsystem (OSS)

The operation subsystem contains all the functions necessary for network
operation and maintenance. It consists of the following:
Operation and Maintenance Centre (OMC): It supervises all other
network entities. Its functions are traffic monitoring, subscribers, security
management and accounting billing.
Authentication Centre (AuC): It protects against intruders targeting the
air interface. The AuC stores information concerned with security features
such as user authentication and encryption. The AuC is related to the HLR.

Equipment Identity Register (EIR): It is essentially a database that is

used to track handsets using the IMEI. It helps to block calls from stolen,
unauthorized, or defective mobiles.

2.7.3 GSM Security

Security in GSM is broadly supported at three levels: operator’s level,

customer’s level and system level. These three levels help oversee aspects
such as correct billing to the customer, avoiding fraud, protecting services,
and ensuring anonymity. The following are a few important features
associated with providing security in GSM networks.

Authentication
The purpose of authentication is to protect the network against unauthorized
use. In the GSM context, it helps protect the GSM subscribers by denying
the possibility for intruders to impersonate authorized users. A GSM
network operator can verify the identity of the subscriber, making it highly
improbable to clone someone else’s mobile phone identity.
Authentication can be achieved in a simple way by using a password
such as Personal Identification Number (PIN). This method is not very
secure in GSM networks as an attacker can “listen” the PIN and easily
break the code.

Confidentiality
A GSM network protects voice, data and sensitive signalling information
(e.g. dialed digits) against eavesdropping on the radio path. Confidentiality
of subscriber-dialled information in the GSM network is achieved by
using encryption techniques prescribed by the GSM designers. Data on
the radio path is encrypted between the Mobile Equipment (ME) and
the BTS which protects user traffic and sensitive signalling data against
eavesdropping.
 Introduction to Mobile Computing and Wireless Networking 41

Anonymity
A GSM network protects against someone tracking the location of a user
or identifying calls made to (or from) the user by eavesdropping on the
radio path. The anonymity of the subscriber on the radio access link in
the GSM network is achieved by allocating Temporary Mobile Subscriber
Identity (TMSIs) instead of permanent identities. This helps to protect
against tracking a user’s location and obtaining information about a user’s
calling pattern.

2.8 General Packet Radio Service (GPRS)

GPRS when integrated with GSM, significantly improves and simplifies

Internet access. It transfers data packets from GSM mobile stations to
external packet data networks (PDNs). Packets can be directly routed from
the GPRS mobile stations to packet switched networks making it easy to
connect to the Internet.
GSM uses a billing system based on the time (duration) of connection,
whereas GPRS uses a billing system based on the amount of transmitted
data rather than the duration of the connection. So, users can remain
continuously connected to the system, and yet get charged only for the
amount of transmitted data.

2.8.1 GPRS Services

GPRS offers end-to-end packet-switched data transfer services which can

be categorized into the following two types:
(i) Point-to-Point (PTP) service
(ii) Point-to-Multipoint (PTM) service.
The PTP service is between two users and can either be connectionless or
connection-oriented. The PTM is a data transfer service from one user to
multiple users. Again, there are two types of PTM services. One is multicast
PTM where the data packets are broadcast in a certain area and the other is
group call PTM where the data packets are addressed to a group of users.

2.8.2 GPRS Architecture

GPRS architecture introduces two new network elements, called GPRS
Support Node (GSN) and the Gateway GPRS Support Node (GGSN).
A GSN is essentially a router. All GSNs are integrated into a standard
GSM architecture. The GGSN is the interworking unit between the GPRS
network and the external packet data network (PDN). The GGSN contains
42 Fundamentals of Mobile Computing

routing information for GPRS users, performs address connection and

tunnells data to a user through encapsulation. In Fig. 2.10, the GGSN is
connected to an external network and it transfers packets to the SGSN
through an IP-based GPRS backbone network.

Figure 2.10 GPRS architecture reference model.

As shown in Fig. 2.10, SGSN (Serving GPRS Support Node) helps

support MS. The SGSN is connected to BSC through frame relay and it is
at the same hierarchy level as the MSC. The GPRS Register (GR) is a part
of HLR which stores all the relevant GPRS data. In a part of HLR which
stores all the relevant data of GPRS in a mobile IP network, GGSN and
SGSNs can be compared with home agent and foreign agent respectively.
The data packets are transmitted to the BSS and finally to the MS through
the GGSN and SGSN. The MSC as we have already discussed is responsible
for data transport in the traditional circuit-switched GSM.

2.9 Universal Mobile Telecommunications System (UMTS)

CDMA2000 and UMTS were developed separately and are two separate
ITU approved 3G standards. In these networks, coverage is provided by
a combination of various cell sizes, ranging from “in building” pico cells
to global cells provided by satellites, giving service to the remote regions
of the world.
The UMTS was developed mainly for countries with GSM networks,
and it is expected that all GSM networks will be upgraded to UMTS
networks. Because it is a new technology, a whole new radio access
network has to be built. An important advantage of UMTS is that it gives
significantly enhanced capacities to operators.
 Introduction to Mobile Computing and Wireless Networking 43

The UMTS specification has been designed so that the UMTS systems
are compatible with GSM networks. Therefore, the UMTS networks can
easily work with any existing GSM/GPRS network. The UMTS systems use
different frequency bands, so the BTSs do not interfere with each other.
Let us now discuss the dissimilarities between these networks. The
UMTS networks are different from the 2G networks in the following
respects:
Higher speech quality: In addition to speech traffic, the UMTS supports
the advanced data and information services and can be called a true
multimedia network.
Higher data rate: The UMTS supports 2 Mbps data rate, which is much
higher than that supported by the 2G mobile systems.

Virtual home environment (VHE): A user roaming from his network to

other UMTS networks will not feel any discontinuity or service difference,
thus giving a “feeling” of being in the home network. In contrast, in a 2G
network, a user is registered to a visitor location and is also charged a
roaming overhead.

2.9.1 UMTS Network Architecture

The UMTS network architecture can be divided into three main elements:

User Equipment (UE): The User Equipment (UE) is the name by which
a cell phone is referred to. The new name was chosen because of the
considerably greater functionality that the UE incorporates compared to a
cell phone. It can be thought of as both a mobile phone used for talking
and a data terminal attached to a computer with no voice capability.
Radio Network Subsystem (RNS): The RNS is the equivalent of the Base
Station Subsystem (BSS) in GSM. It provides and manages the wireless
interface for the overall network.
Core Network: The core network is the equivalent of the GSM Network
Switching Subsystem (NSS).

2.10 Mobile Phone and Human Body

Extensive research has been conducted to study the possible health effects
of continuous exposure to low intensity RF fields produced by mobile
phones. The overall evidence suggests that mobile phone usage of less than
10 years does not pose any increased risk of brain tumour. The effect of
still longer use is unclear due to non-availability of data. Any conclusion
44 Fundamentals of Mobile Computing

therefore is uncertain and tentative. From the available data, however, it

does appear that there is no increased risk for brain tumours in long-term
users as well, with the exception of acoustic neuroma (see Box 2.2) for
which there is limited evidence of a weak association. Available studies
suggest that self-reported symptoms are not correlated to an acute exposure
to RF fields, but considering that the studies are carried out over a rather
limited duration, it is very difficult to draw any firm conclusions. Currently
available studies on neurological effects and reproductive effects have not
indicated any health risks at exposure levels below guidelines.

BOX 2.2 Acoustic neuroma

Acoustic neuroma is a non-cancerous tumour that develops on the nerve that
connects the ear to the brain. The tumour usually grows slowly. As it grows,
it presses against the hearing and balance nerves. At first, there may be no
symptoms or mild symptoms. They can include the following:
• Loss of hearing on one side
• Ringing in ears
• Dizziness and balance problems

SUMMARY

Mobile computing is an umbrella term used to describe technologies

that enable people to access network services—anyplace, anytime, and
anywhere. The term ubiquitous computing and nomadic computing are
often used synonymously with mobile computing. Mobile computing
essentially involves accessing and manipulating information over a
wireless network using a hand-held device. The challenges in designing
effective mobile computing solutions include providing acceptable service
in the presence of low available bandwidth, intermittent disconnections,
poor security, and difficulty in assigning fixed addresses to hosts. Unlike
their wired counterparts, the design of software for mobile devices must
consider resource limitation, battery power and display size. Consequently,
new hardware and software techniques must be developed. For example,
applications need to be highly optimized for space, in order to fit in
the limited memory on the mobile devices. For Internet enabled devices,
the good old TCP/IP stack is not very suitable, it is computationally
expensive for hand-held devices, does not make efficient use of the available
bandwidth and is not optimized for low power consumption.
Digital cellular standards meet the current requirements of voice
communications and are being upgraded to meet the future demands in
mobile multimedia applications. The 3G mobile networks represent an
evolution in terms of capacity, data speeds and new service capabilities
from second generation mobile networks to provide an integrated solution
for mobile voice and data with wide area coverage.
 Introduction to Mobile Computing and Wireless Networking 45

FURTHER READINGS

Abdelsalam Helal, Bert Haskell, J. Carter, R. Brice, D. Woelk and

M. Rusinkiewicz, Anywhere Computing Concepts and Technology, Kluwer
Academic Publishers, September 1999.
Adelstein, Gupta, Rechard III, and Schwiebert, Fundamentals of Mobile and
Pervasive Computing, TMH.
Akyildiz, I.F. and J.S.M. Ho, “ Dynamic Mobile User Location Update for
Wireless PCS Networks”, ACM/Baltzer Wireless Networks Journal, 1995.
Arkady Zaslavsky and Zahir Tari, “Mobile Computing Overview and
Current Status”, Australian Computer Society, 1998.
Christian Bettstetter, Hans-Jorg Vogel, and Jorg Eberspacher, “GSM Phase 2
+ GPRS Architecture, Protocol and Air Interface”, IEEE Communication
Survey, 1999, Vol. 2.
Jan A. Audestad, “Network Aspect of GSM System”, EURCCON, June 1988.
Max Stepanov, “GSM Features and Security”, July 2004.
Rappaport, T., Wireless Communication and Principles, Pearson Education.
Tuan Huynh and Hoang Nguyen, “Overview of GSM and GSM Security”,
2003.

EXERCISES
1. State true or false against each of the following statements. Give
appropriate reasons for your answer.
(i) Mobile computing and wireless networking are synonymous
terms.
(ii) The Bluetooth technology is an example of an infrastructure-
less network.
(iii) WLAN is an example of an infrastructure-less network.
(iv) A sensor network is essentially based on a single hop wireless
communication.
(v) 2G cellular phones support electronic mail services.
(vi) In the Global Packet Radio Service (GPRS), the customer is
charged for Internet access based on connection time rather
than on total information download.
(vii) GSM stands for Group of Special Mobiles.
(viii) UMTS networks can easily work with the existing GSM/GPRS
networks.
2. What is mobile computing? Mention at least three applications of
mobile computing.
46 Fundamentals of Mobile Computing

3. Distinguish between mobile computing and wireless networking.

4. Give an overview of the working of current mobile cellular phones.
Briefly explain the distinguishing features of various generations of
wireless cellular networks.
5. Briefly explain how the mobile cellular communication has evolved
over different generations of technology.
6. What do you understand by 2.5G? Mention a few characteristic
features of this technology. How is it different from 2G and 3G
technologies?
7. Distinguish between infrastructure-based networks and infrastruc-
ture-less networks with the help of suitable schematic diagrams.
8. Briefly write about the Bluetooth technology. Give at least two
examples of its use.
9. Explain the architecture of a mobile computing environment. Define
the functions of the presentation tier, application tier and data tier
of mobile computing environment.
10. Briefly discuss the important functional differences between 1G, 2G,
and 3G cellular networks.
11. Compare 1G and 2G cellular wireless communication technologies.
12. Is 3G cellular wireless technology superior to 2G technology? Justify
your answer.
13. Explain how a GSM network provides security to the customers.
14. What are the advantages of GPRS over GSM?
15. What is UMTS? Describe the functions of HLR and VLR in call
routing and roaming?
16. Identify at least three similarities and three dissimilarities between
a GMS network and a UMTS network.
17. What do you mean by Virtual Home Environment (VHE)? Explain
how VHE is realized in 3G networks.
18. Do mobile phones affect the human body negatively? Explain your
answer.
19. Discuss the advantages and disadvantages of supporting TCP/IP in
a mobile computing network.
20. Identify the main reasons as to why a mobile handset is compact
and lightweight and yet provides a large number of features such
as roaming, camera, audio and video play and record, Internet
browsing, etc., while the traditional landline phone handsets are
bulky and provide only limited features.
21. What is the difference between analog and digital transmissions
in the context of mobile communications? What are their relative
advantages?
 MAC Protocols
3
I n a wireless network, multiple nodes may contend to transmit on the
same shared channel at the same time. In this situation, the transmitted
data would get garbled unless a suitable medium access arbitration scheme
is deployed. Usually, it is the responsibility of the medium access control
(MAC) protocol to perform this task. The MAC protocol is a sublayer of the
data link layer protocol and it directly invokes the physical layer protocol.
The primary responsibility of a MAC protocol is to enforce discipline
in the access of a shared channel when multiple nodes contend to access
that channel. At the same time, two other objectives of any MAC protocol
are maximization of the utilization of the channel and minimization of
average latency of transmission. However, a MAC protocol must be fair
and ensure that no node has to wait for an unduly long time, before it is
allowed to transmit. In the following, we identify the various characteristics
desirable of any MAC protocol.

3.1 Properties Required of MAC Protocols

The design of a MAC protocol depends upon the specific environment in

which it would be used and the specific design requirements to be met.
In spite of the wide variations in the characteristics of different protocols,
however, in a general sense a good MAC protocol needs to possess the
following features:
• It should implement some rules that help to enforce discipline when
multiple nodes contend for a shared channel.
• It should help maximize the utilization of the channel.
• Channel allocation needs to be fair. No node should be discriminated
against at any time and made to wait for an unduly long time for
transmission.
47
48 Fundamentals of Mobile Computing

• It should be capable of supporting several types of traffic having

different maximum and average bit rates.
• It should be robust in the face of equipment failures and changing
network conditions.
Many MAC layer protocols for wireless networks have already been
proposed, standardized, and are in use. Also, many other protocols that
work with improved efficiency or overcome some problems in specific
wireless environments, are being proposed by researchers and practitioners.
At present, IEEE 802.11 has emerged as a popular and standard MAC
protocol for wireless networks. IEEE 802.11-based network cards and
routers are available in the market that can be used to inexpensively and
easily set up wireless LANs (commonly referred to as Wi-fi hotspots). As
we discussed in Chapter 1, wireless networks can be divided mainly into
two categories: (a) infrastructure-based wireless networks that include
the WLANs, and (b) infrastructure-less wireless networks that include
the mobile ad hoc networks (MANETs). Though the MAC protocols for
these two environments have many things in common, MAC protocols for
Infrastructure-less networks are surprisingly much more complex as they
have to address certain additional problems that arise in the infrastructure-
less environments. We discuss MAC protocols for both these environments
in this chapter.
In Section 3.2, we first discuss a few basic issues concerning the MAC
protocols in a wireless network. In Section 3.3, we discuss a taxonomy of
MAC protocols for wireless networks. In the three subsequent sections,
we discuss the three important categories of MAC protocols: the fixed
assignment, the random access, and the reservation-based protocols.
Finally, we discuss the MAC protocols that have been specifically designed
for ad hoc networks.

3.2 Wireless MAC Protocols: Some Issues

A MAC protocol in a wireless medium is much more complex than its wired
counterpart. First, a collision detection scheme is difficult to implement
in a wireless environment, since collisions are hard to be detected by
the transmitting nodes. Also, in infrastructure-less networks, the issue of
hidden and exposed terminals make a MAC protocol extremely inefficient
unless special care is taken to overcome these problems. We elaborate the
hidden and exposed terminal problems in the following:

3.2.1 The Hidden and Exposed Terminal Problems in

an Infrastructure-less Network

The hidden terminal problem arises when at least three nodes (A, B, and C), as
shown in Fig. 3.1, communicate among each other. As shown in this figure,
 MAC Protocols 49

B is in the radio range of A, and B is also within the radio range of C.

However, the nodes A and C are not in the radio range of each other.
Note that if both A and C start to transmit to B at the same time, the data
received at node B would get garbled. Such a situation can arise because A
and C are “hidden” from each other, because they are outside each other’s
transmission range. In this situation, when one node starts to sense the
medium before transmission, it cannot sense that the other node is also
transmitting. This creates a very difficult and important arbitration problem
that a MAC protocol needs to resolve.

Figure 3.1 Hidden terminal problem.

A related problem called exposed terminal could arise in a scenario such

as that depicted in Fig. 3.2. MAC protocols usually inhibit transmission
when transmission from another terminal is detected. As a result, node
A will not be able to transmit to any node when B is transmitting to C.
On the other hand, had A transmitted to D, it would have been received
correctly by D and B’s transmission would have also been correctly received
at C. The problem arose only because A and B are within each other’s
transmission range, though the destination nodes are in the transmission
range of only one of the nodes. In other words, the problem occurs because
A is exposed to B’s transmission. The overall effect of this problem is that
it leads to inefficient spectrum usage as well as unnecessary transmission
delays unless these are carefully addressed by a wireless MAC protocol.

A B C

Figure 3.2 Exposed terminal problem.

50 Fundamentals of Mobile Computing

3.3 A Taxonomy of MAC Protocols

A large number of MAC protocols have been proposed. These MAC
protocols can be broadly divided into the following three categories:
(i) Fixed assignment schemes
(ii) Random assignment schemes
(iii) Reservation-based schemes
The fixed assignment schemes are usually called circuit-switched
schemes. In the fixed assignment schemes, the resources required for a
call are assigned for the entire duration of the call. On the other hand, the
random assignment schemes and the reservation schemes are called packet-
switched schemes. The random assignment schemes are comparable to the
connection-less packet-switching schemes. In this, no resource reservations
are made, the nodes simply start to transmit as soon as they have a packet
to send. In the reservation schemes, a node makes explicit reservation
of the channel for an entire call before transmitting. This is analogous
to a connection-based packet-switching scheme. The reservation-based
MAC schemes are suitable to handle calls with widely varying traffic
characteristics. In the following sections, we discuss these three categories
of MAC protocols in some more detail.

3.4 Fixed Assignment Schemes

A few important categories of fixed assignment MAC protocols are the
following:
• Frequency Division Multiple Access (FDMA)
• Time Division Multiple Access (TDMA)
• Code Division Multiple Access (CDMA)
We briefly discuss these techniques in the following subsections.

BOX 3.1 An analogy to the fixed assignment solution to the multiple

access issues of a shared medium
An analogy may be drawn to the fixed assignment solution to the multiple
access issues of a shared medium in the following way: Consider a students’
common room (channel) in which many students want to communicate with
each other. If the students want to avoid cross-talk in the ongoing process,
then either the students could take turns in speaking (i.e. time division), or
they could speak at different pitches (i.e. frequency division), or they could
speak in different languages (i.e. code division). The last analogy captures the
essence of CDMA, when the students who are speaking the same language
understand each other, but the rest of the students cannot. In CDMA, each
communicating pair shares a decryption code using which lets them understand
only the communication between them. In this case many codes occupy the
same channel, but only the users who share a specific code will be able to
understand each other.
 MAC Protocols 51

3.4.1 Frequency Division Multiple Access (FDMA)

In FDMA, the available bandwidth (frequency range) is divided into many
narrower frequency bands called channels. Figure 3.3 shows a division
of the existing bandwidth into many channels (showns as Ch 1, Ch 2,
etc.). For full duplex communication to take place, each user is allocated
a forward link (channel) for communicating from it (mobile handset) to
the base station (BS), and a reverse channel for communicating from the
BS to it. Thus, each user making a call is allocated two unique frequency
bands (channels), one for transmitting and the other for receiving signals
during the call. Obviously, when a call is underway, no other user would
be allocated the same frequency band to make a call. Unused transmission
time in a frequency band that occurs when the allocated caller pauses
between transmissions, or when no user is allocated a band, goes idle and
is wasted. FDMA, therefore, does not achieve a high channel utilization.

Figure 3.3 Channels in Frequency Division Multiple Access (FDMA) scheme.

3.4.2 Time Division Multiple Access (TDMA)

TDMA is an access method in which multiple nodes are allotted different
time slots to access the same physical channel. That is, the timeline is
divided into fixed-sized time slots and these are divided among multiple
nodes who can transmit. Note that in this case, all sources use the same
channel, but take turns in transmitting. Figure 3.4 shows the situation
where time slots are allocated to users in a round robin manner, with
each user being assigned one time slot per frame. See Box 3.2. Obviously,
unused time slots go idle, leading to low channel utilization.

Figure 3.4 Channels in Time Division Multiple Access (TDMA) scheme.

BOX 3.2 TDMA scheme

In TDMA, each user of the channel owns the channel for exclusive use for one
time slot at a time in a round robin fashion.
52 Fundamentals of Mobile Computing

3.4.3 Code Division Multiple Access (CDMA)

In CDMA, multiple users are allotted different codes that consist of

sequences of 0 and 1 to access the same channel. As shown in Fig. 3.5, a
special coding scheme is used that allows signals from multiple users to be
multiplexed over the same physical channel. As shown in the figure, three
different users who have been assigned separate codes are multiplexed on
the same physical channel.

Figure 3.5 Schematic of operation of Code Division Multiple Access (CDMA).

In the following, we elaborate the CDMA technology. In CDMA,

multiple users use the same frequency at the same time and no time
scheduling is applied. All the senders send signals simultaneously through
a common medium. The bandwidth of this medium is much larger than
the space that would be allocated to each packet transmission during
FDMA and the signals can be distinguished from each other by means
of a special coding scheme that is used. This is done with the help of a
frequency spreading code known as the m-bit pseudo-noise (PN) code
sequence. Using m bits, 2m – 1 different codes can be obtained. From these
codes, each user will use only one code.

BOX 3.3 How to distinguish transmission from different nodes

Two vectors are said to be orthogonal if their inner product = 0. Let p and q
be two vectors and suppose p = (2, 5, 0) and q = (0, 0, 17), then the inner
product of pq = (20+50+017) = 0.
For good autocorrelation, binary 0 is represented as –1 and binary 1 is
represented as +1. Let the binary sequence be 1001, then the representation
sequence is +1–1–1+1.
 MAC Protocols 53

It is possible to distinguish transmissions from different nodes by

ensuring some properties on the codes. A code for a user should be
orthogonal (that is, non-interfering) to the codes assigned to other nodes.
The term “orthogonal” means that the vector inner product is zero, and
good autocorrelation uses the bipolar notation where a code sequence of
binary 0 is represented as –1 and binary 1 is represented as +1. See Box 3.3.
On the receiving end, only the same PN sequence is able to demodulate
the signal to successfully convert the input data.

BOX 3.4 Pseudorandom sequence generator

To generate a series of pseudorandom numbers, a seed (or starting point) is
required. Based on the selected seed, the next number can be generated using a
deterministic mathematical transformation or can be generated probabilistically.
In CDMA, a code actually denotes a starting point (seed) for a pseudorandom
sequence generator (PRSG). PRSG generates a series of bits at a frequency
which is much higher than the actual user data (such as digitized voice). These
bits are XORd with the user data and subsequently the results are transmitted.
This occurs in the case of multiple transmitters.
If someone listens to this signal with the help of a suitable wideband
receiver, the person will hear something similar to what is produced by random
noise. All the other users who are on the same frequency will send a similar
signal, but with a different PRSG seed. So these apparent random noises will
all coexist in the same band of frequencies, but would not interfere with each
other. This is due to the reason that the exact frequency of any transmitter at
any instant (which is in effect determined by the seed) is almost always unique.
Error correction takes care of occasional bit errors.
The receiver is aware of the PRSG starting point for each transmitter. It
hears just one of the transmitters by correlating the noise it receives, against
its own PRSG, which is also running with the same seed. It is slightly similar
to FDMA in this sense, but the difference is that the transmitters do not
stay on one frequency. They hop around many times per bit of user data.
The pseudorandom sequence determines this hopping, rather than a fixed
assignment to each transmitter.

For simplicity, we assume that all nodes transmit on the same frequency
at the same time using the entire bandwidth of the transmission channel.
Each sender has a unique random number key, and the sender XORs the
signal with this random number key. The receiver can “tune” into this
signal if it knows the pseudorandom number. Consider an example, where
X, Y are the transmitters and Z is a receiver. Sender X_data = 1 and X_Key
= (010011). Its autocorrelation representation is (–1, +1, –1, –1, +1, +1). The
signal to be calculated at sender X is Xs = X_data  X_key = +1X_key =
(–1, +1, –1, –1, +1, +1). Similarly, sender Y_data = 0 and Y_key = (110101)
and the signal to be sent at Y is Ys = –1Y_key = –1(+1, +1, –1, +1, –1,
+1) = (–1, –1, +1, –1, +1, –1). The signal received by receiver Z is Xs + Ys
= (–1, +1, –1, –1, +1, +1) + (–1, –1, +1, –1, +1, –1) = (–2, 0, 0, –2, +2, 0).
At the receiver, in order to receive the data sent by sender X, the signal
54 Fundamentals of Mobile Computing

Z is dispread. So now if Z wants to get information of sender X data,

then ZX_key = (–2, 0, 0, –2, +2, 0)(–1, +1, –1, –1, +1, +1) = 2 + 0 + 0 +
2 + 2 + 0 = 6 > 0 (positive), that is the original bit was a 1. Similarly, the
information of sender Y data may be obtained as ZY_key = (–2, 0, 0, –2,
+2, 0)(+ 1, +1, –1, +1, –1, +1) = –2 + 0 + 0 – 2 – 2 + 0 = – 6 < 0 (negative).
So the Y data original bit was a 0.

3.5 Random Assignment Schemes

There are a number of random assignment schemes that are used in MAC
protocols. A few important ones are the following:
• ALOHA
• Slotted ALOHA
• CSMA
• CSMA/CD
• CSMA/CA

3.5.1 ALOHA Scheme

It is a simple communication scheme that was developed at the university

of Hawaii. The basic (also called pure) ALOHA scheme, is a simple
protocol. If a node has data to send, it begins to transmit. Note that the
first step implies that Pure ALOHA does not check whether the channel is
busy before transmitting. If the frame successfully reaches the destination
(receiver), the next frame is sent. If the frame fails to be received at the
destination, it is sent again. The simple ALOHA scheme works acceptably,
when the chances of contention are small (i.e., when a small number
of senders send data infrequently). However, the collisions can become
unacceptably high if the number of contenders for transmission is high.
An improvement over the pure ALOHA scheme is the slotted ALOHA.
In the slotted ALOHA scheme, the chances of collisions are attempted to
be reduced by enforcing the following restrictions. The time is divided
into equal-sized slots in which a packet can be sent. Thus, the size of the
packet is restricted. A node wanting to send a packet, can start to do so
only at the beginning of a slot. The slotted ALOHA system employs beacon
signals that are sent at precise intervals that mark the beginning of a slot,
at which point the nodes having data to send can start to transmit. Again,
this protocol does not work very well if the number of stations contending
to send data is high. In such cases, the CSMA scheme (described next)
works better.
 MAC Protocols 55

3.5.2 The CSMA Scheme

A popular MAC arbitration technique is the Carrier Sense Multiple Access

(CSMA). In this technique, a node senses the medium before starting to
transmit. If it senses that some transmission is already underway, it defers
its transmission. Two popular extensions of the basic CSMA technique are
the collision detection (CSMA/CD) and the collision avoidance (CSMA/
CA) techniques.
Unlike that in a wired network, in a wireless network the CSMA/CD
technique does not work very well. In the CSMA/CD technique, the sender
starts to transmit if it senses the channel to be free. But, even if it senses
the channel to be free, there can be a collision (why?) during transmission.
In a wired network, the implementation of a collision detection scheme is
simple. However, in a wireless network it is very difficult for a transmitting
node to detect a collision, since any received signal from other nodes
would be too feeble compared to its own signal and can easily be masked
by noise. As a result, a transmitting node would continue to transmit the
frame, and only the destination node would notice the corrupted frame
after it computes the checksum. This leads to retransmissions and severe
wastage of channel utilization. In contrast, in a wired network when a node
detects a collision, it immediately stops transmitting, thereby minimizing
channel wastage.
In a wireless network, a collision avoidance scheme works much better
compared to a collision detection-based scheme. A collision avoidance
scheme is based on the idea that it is necessary to prevent collisions at the
moment they are most likely to occur, that is, when the bus is released
after a packet transmission. We explain the reason for this in the following.
During the time a node is transmitting on the channel, several nodes might
be wanting to transmit. These nodes would be monitoring the channel
and waiting for it to become free. The moment the transmitting node
completes its transmission, these waiting nodes would sense the channel
to be free, and would all start transmitting at the same time. To overcome
such collisions, in the collision avoidance scheme, all nodes are forced to
wait for a random time and then sense the medium again, before starting
their transmission. If the medium is sensed to be busy, a node waiting to
transmit waits for a further random amount of time and so on. Thus, the
chance of two nodes starting to transmit at the same time would be greatly
reduced.

3.6 Reservation-based Schemes

A basic form of the reservation scheme is the RTS/CTS scheme. In an

RTS/CTS scheme, a sender transmits an RTS (Ready to Send) packet to
the receiver before the actual data transmission. On receiving this, the
56 Fundamentals of Mobile Computing

receiver sends a CTS (Clear to Send) packet, and the actual data transfer
commences only after that. When the other nodes sharing the medium
sense the CTS packet, they refrain from transmitting until the transmission
from the sending node is complete.
In a contention-based MAC protocol, a node wanting to send a message
first reserves the medium by using an appropriate control message. For
example, reservation of the medium can be achieved by transmitting a
“Ready To Send” (RTS) message and the corresponding destination node
accepting this request answers with a “Clear To Send” (CTS) message. Every
node that hears the RTS and CTS messages defers its transmission during
the specified time period in order to avoid a collision. A few examples of
RTS-CTS based MAC protocols are MACA, MACAW, MACA-BI, PAMAS,
DBTMA, MARCH, S-MAC protocols which have specifically been designed
for sensor networks. In the following, we discuss MACA as a representative
protocol belonging to this category of protocols.

3.6.1 MACA

MACA stands for Multiple Access Collision Avoidance. MACA solves the
hidden/exposed terminal problems by regulating the transmitter power.
A node running MACA requests to use the medium by sending an RTS
to the receiver. Since radio signals propagate omni-directionally, every
terminal within the sender’s radio range will hear this and then refrain
from transmitting. As soon as the receiver is ready to receive data, it
responds with a CTS.
Figure 3.6 schematically shows how MACA avoids the hidden terminal
problem. Before the start of its transmission, it sends a Request To Send
(RTS). B receives the RTS that contains the sender’s name and the receiver’s
name, as well as the length of the future transmission. In response to the
RTS, an acknowledgment from B is triggered indicating Clear To Send
(CTS). The CTS contains the names of the sender and receiver, and the
length of the planned transmission. This CTS is heard by C and the medium
is reserved for use by A for the duration of the transmission.

Figure 3.6 Hidden terminal solution in MACA.

 MAC Protocols 57

On receipt of a CTS from B, C refrains from transmitting anything for

the time indicated in the CTS. Thus a collision cannot occur at B during
data transmission, and the hidden terminal problem is solved.
Though this is a collision avoidance protocol, a collision can occur
during the sending of an RTS. Both A and C could send an RTS at same
time. But an RTS occurs over a very small duration compared to the
duration of data transmission. Thus the probability of collision remains
much less. B resolves this contention problem by acknowledging only one
station in the CTS. No transmission occurs without an appropriate CTS.
Figure 3.7 schematically shows how the exposed terminal problem is
solved in MACA. Assume that B needs to transmit to A. B has to transmit
an RTS first as shown in Fig. 3.7. The RTS would contain the names of the
receiver (A) and the sender (B). C does not act in response to this message
as it is not the receiver, but A responds with a CTS. C does not receive
this CTS and concludes that A is outside the detection range. Thus C can
start its transmission assuming that no collision would occur at A.

Figure 3.7 Exposed terminal solution in MACA.

3.7 The 802.11 MAC Standard

The IEEE 802.11 is the most widely used standard for WLANs today.
Low cost WLAN cards can be easily purchased from the market to set up
a WLAN. The IEEE 802.11 standard defines the functional aspects of the
medium access control (MAC) sublayer. The IEEE 802.11 defines separate
standards for infrastructure-based and ad hoc networks. The network
interface cards can be set to work in either of these two modes.

3.7.1 Infrastructure-based Mode

This mode of usage is widely used and helps provide Wi-Fi hot spots in a
campus to access the Internet. It is based on the CSMA/CA protocol due
to the obvious advantages of this protocol over the CSMA/CD schemes
58 Fundamentals of Mobile Computing

in a wireless environment, which we discussed in Section 3.5. In CSMA,

a station that wants to transmit will first have to listen to the channel for
a pre-determined period of time for checking if there is any transmission
activity occurring on the channel. The station would transmit if it senses the
channel to be idle. The transmission by the channel is deferred, if it senses
the channel to be busy. CSMA can be considered a greedy protocol since
a node trying to transmit, puts the packet out on the channel the moment
it finds that the channel is idle. CSMA uses collision avoidance to improve
the system’s performance by trying to be less greedy. In case, the channel
is sensed to be busy when it tries to transmit, then the transmission will
be deferred for a “random” interval, thereby reducing the probability of
further collisions on the channel. CSMA/CA is used in the 802.11-based
wireless LANs.
The RTS/CTS (Request to Send/Clear to Send) is an optional mechanism
used by the 802.11 to reduce frame collisions caused by the hidden node
problem. As we discussed, a Request to Send (RTS) packet is sent by a
sender and a Clear to Send (CTS) packet is sent by the intended receiver.
The CTS alerts all the nodes within the range of both the sender or the
receiver to keep quiet for the duration of the transmission.
The IEEE 802.11 has been extensively used because of its installation
and operational simplicity and low cost. However, it should be borne in
mind that it can be used only for a single-hop communication. That is, all
nodes can communicate as long as they are in the transmission range of a
single access point. This limitation can be overcome by multi-hop ad hoc
networking. The 802.11a operates at 54 Mbps. Note that in 802.11, since
the transmitter power is constant and limited, the data rate decreases when
the transmission range increases.

3.8 MAC Protocols for Ad Hoc Networks

Ad hoc networks are infrastructure-less, self-organizing networks of mobile

computers typically carried around by people. These computers are equipped
with radio-enabled network interface cards for communication purposes.
Every node can communicate directly with the other neighbouring nodes
who can “hear” its transmissions. The computers detect other computers
in their neighbourhood, dynamically forming links that set up the network.
A link is also called one “hop”. In addition to communicating with other
computers in the neighbourhood, a computer can communicate with remote
computers not in the vicinity by letting intermediate computers relay
information, thereby forming a multi-hop path to the required destination
computer. In forming a multi-hop path to a destination computer, a
complexity that arises is on account of the mobility of the computers.
Thus, a multi-hop path can become stale (that is, invalid) after some time,
or might even break while a communication is still going on. Thus, no fixed
infrastructure is required for setting up an ad hoc network.
 MAC Protocols 59

Wireless multi-hop ad hoc networks are comprised of lightweight

mobile radio frequency bands which are a scarce resource and therefore
these are shared among all hosts. Medium Access Control (MAC) protocols
for ad hoc networks can be either centralized or distributed. For instance,
Bluetooth uses a centralized scheme where the master of a network assigns
turns for transmission to slaves. The IEEE 802.11 is a MAC protocol that
can utilize both schemes, but in ad hoc networks the distributed scheme
is used. In this scheme, nodes contend for transmission turns, but if more
than one node wants to transmit at the same time, a random node is chosen
for transmission. The important categories of MAC protocols for MANET
are discussed below.
There are essentially two broad categories of MAC protocols for
ad hoc networks. The first category controls shared medium access by
letting terminals compete asynchronously. In this protocol, every node
is free to transmit any time. But upon detection of a collision, the access
arbitration policy is invoked to avoid the collision. The second type of
MAC protocol divides the medium into channels so that each competing
node uses a different channel, thereby avoiding collisions. One way to
achieve this is by dividing the transmission time (slots), and inserting a
frequency band between terminals and requiring them to synchronize in
order to ensure that they never use the same slot, frequency or code. In
ad hoc networks, only the terminals that are present in the neighbourhood
have to be synchronized. This is because the non-neighbour terminals
cannot interfere with each other. Garcia et. al have named the first type
“contention-based” protocols and the second type “contention-free”
protocols. The RTS-CTS based schemes are contention-based protocols and
the multiple access protocols (FDMA/TDMA/CDMA) are contention-free
protocols. In Fig. 3.8, we have shown this classification of the protocols.

Figure 3.8 A classification of MAC protocols for ad hoc networks.

In another category of protocols called fixed-assignment schemes, no

contention is involved. These protocols use techniques such as FDMA,TDMA
and CDMA. Protocols using this kind of medium access methods are
60 Fundamentals of Mobile Computing

contention-free because every node can transmit without worrying about

collisions.
As discussed, the RTS-CTS based schemes are contention-based.
Contention-based protocols incur additional overhead due to the
reservation messages. In a contention-based protocol, the delay that a
message undergoes is not deterministic, since it would depend on whether
the other nodes are trying to send messages at that time or not. Further, in
a contention-based protocol, as the network traffic increases, the number
of contentions increase. This in turn causes the throughput to degrade
dramatically. A positive aspect of the contention-based protocols is their
simplicity and relative easy of implementation.
Contention-free protocols overcome many of the shortcomings of the
contention-based protocols. Packets in a contention-free protocol are not
delayed on account of collisions. However, contention-free protocols are
difficult to implement in a multi-hop ad hoc network. We therefore do not
discuss contention-free ad hoc network protocols. In general, contention-free
protocols for wireless networks can be both centralized and distributed. For
instance, Bluetooth piconet uses a contention-free centralized MAC protocol
in which time slots are divided among terminals by the master. Also, cell-
based, single-hop, mobile phone networks use centralized contention-free
protocols. Interestingly, ad hoc networks that have controlling masters use
centralized contention-free protocols. Therefore, Bluetooth scatternet would
also qualify as a contention-free centralized MAC protocol.
Multi-access protocols do not need a centralized master but the
neighbouring nodes must agree on the specific resource (time, frequency, or
code) to use, so that there is no conflict with the other nodes. The problem
of assigning resource (time, frequency, code) to different pairs of nodes in
a multi-access protocol, is equivalent to the well-studied problem in graph
theory called the edge colouring problem in which no two edges (links) can
be assigned the same colour (out of a finite set of colours) if they share a
common vertex. We will not discuss the details of any MAC protocol and
restrict our discussions to the conceptual level only.
Much of the research done for MANET may also apply to sensor
networks since both operate as multi-hop wireless networks with power
constraints. MANET, however, focuses on device mobility, while sensor
networks normally have limited or no mobility. Long studied problems
in wireless networks, such as the hidden terminal and exposed node
problem, also exist in sensor networks, so protocol designers must handle
these issues in addition to the characteristics unique to sensor networks.
Researchers now have the challenge to solve existing problems from
traditional wireless networks under the constraints introduced by the
limited resources available in WSN. In this, nodes define common active/
sleep periods. The active periods are used for communication and the
sleep ones for saving energy. This approach requires that nodes maintain a
certain level of synchronization to keep the active/sleep periods common to
all nodes. During the active periods, nodes contend for the medium using
 MAC Protocols 61

CSMA, IEEE 802.11 DCF, etc. However, the use of common active/sleep
periods may not be suitable for applications with irregular traffic, because
nodes use contention inside active periods, which would be prohibitive
when nodes wake up without communicating, and may cause collisions
when there is high traffic that cannot be absorbed by the initially envisaged
size of the active periods.

SUMMARY

A MAC protocol regulates the use of a shared physical channel among a

set of nodes. It arbitrates among the contending nodes. We discussed the
three important categories of solutions: fixed-assignment, random access,
and reservation-based. Protocol design for mobile ad hoc networks is
much more complex than that for wireless LANs, since nodes in such
networks are constrained by factors such as low power, limited bandwidth,
intermittent link errors, and hidden and exposed terminal problems. The
IEEE 802.11 protocol has been implemented in low cost network cards and
access points, and has become widely accepted for the design of Wireless
LANs to provide Wi-Fi hot spots.

FURTHER READINGS

Deng, J. and Z. Haas, “Dual Busy Tone Multiple Access (DBTMA): A New
Medium Access Control for Packet Radio Networks”, Florence, Italy,
1998.
Dam, T.V. and K. Langendoen, “An Adaptive Energy-efficient MAC
Protocol for Wireless Sensor Networks”, in SenSys ’03, New York,
USA: ACM Press, pp. 171–180, 2003.
Heidemann, J., W. Ye, and D. Estrin, “Medium Access Control with
Co-ordinated Adaptive Sleeping for Wireless Sensor Networks”,
IEEE/ACM Transactions on Networking, 12, No. 3, pp. 493–506, 2004.
Karn, P., “A New Channel Access Method for Packet Radio”, ARRL/CRRL
Amateur Radio 9th Computer Networking Conference, Sept. 22, 1990.
Lichun Bao, J.J. Garcia-Luna-Aceves, “Hybrid Channel Access Scheduling in
Ad hoc Networks”, Proceedings of the 10th IEEE International Conference
on Network Protocols, 2002.
Rhee, I., A. Warrier, M. Aia, and J. Min, “ZMAC: a Hybrid MAC for
Wireless Sensor Networks”, New York, USA: ACM Press, 2005,
pp. 90–101.
62 Fundamentals of Mobile Computing

Rom, R. and M. Sidi, “Multiple Access Protocols—Performance and

Analysis”, Springer-Verlag, New York, 1990.
Singh, S. and C.S. Raghavendra, “PAMAS: Power-Aware Multi-Access
Protocol with Signalling for Ad hoc Networks”, ACM Computer
Communication Review, July 1998.
Sunil Kumar, Vineet S. Raghavan, and Jing Deng, “Medium Access Control
Protocols for Ad hoc Wireless Networks: A Survey”, Ad Hoc Networks,
Elsevier Science, Vol. 4, 2006, pp. 326–358.
Talucci, F., M. Gerla, and L. Fratta, “A Receiver Oriented Access Protocol
for Wireless Multihop Networks”, Proceedings of IEEE PIMRC, 1997.
Toh, C.K., Ad Hoc Wireless Mobile Wireless Networks, Prentice Hall PTR, 2002.
Vaduvur Bharghavan, Alan Demers, Scott Shenker, and Lixia Zhang,
“MACAW: A Media Access Protocol for Wireless LANs”, ACM
SIGCOMM Computer Communication, Review, Vol. 24, Issue 4, October
1994.

EXERCISES
1. What is the role of a MAC protocol? At which ISO/OSI layer does
it operate?
2. What is a hidden terminal? What problem does it create during
wireless communications? Explain your answer using a suitable
schematic diagram.
3. When does the exposed terminal problem arise? Explain your answer
using a suitable example.
4. What are the principal responsibilities of the MAC protocols? How
do MAC protocols for wireless networks differ from those in wired
networks?
5. What are the broad categories of MAC protocols? Name one popular
protocol from each of these categories.
6. Explain the working of a contention-based MAC protocol. Give two
examples of contention-based MAC protocols.
7. What are the different categories of MAC protocols. Identify the
situations under which protocols from one category would be
preferable over the other categories. Explain the working of a
reservation-based MAC protocol.
8. Why are collision-detection based protocols not suitable for wireless
networks?
9. Name one MAC protocol that is used in mobile ad hoc networks.
Briefly explain its working.
 MAC Protocols 63

10. Name one MAC protocol that is used in sensor networks. Briefly
explain its working.
11. What is MACA protocol? In which environment is it suitable? Briefly
explain its working. How does MACA protocol solve the hidden/
exposed terminal problems?
12. What do you mean by a schedule-based MAC protocol? Name a
schedule-based MAC protocol. Briefly explain its working.
13. What is FDMA? Briefly explain its working and at least one of its
important applications.
14. What is TDMA? Briefly explain its working and at least one of its
important applications.
15. Explain the basic scheme of the CDMA protocol. What is the role of
a pseudorandom sequence generator in the working of the CDMA
protocol?
16. Briefly explain the IEEE 802.11 standard and discuss its application.
17. Do you agree with the following statement: “In CSMA/CD protocol,
when two nodes transmit on a shared medium, a collision can occur
only when two nodes start transmitting exactly at the same time
instant.” Explain your answer.
18. Identify the specific reasons as to why the MAC protocols designed
for infrastructure-based wireless networks may not work satisfactorily
in infrastructure-less environments.