. 11
( 132 .)


H. Taub, D. Schilling (1986) Principles of Communication Systems. 2nd edn. McGraw-
Hill, New York.
A. J. Viterbi (1995) CDMA: Principles of Spread Spectrum Communication. Addison-
Wesley, Reading, MA.
A. J. Viterbi (1967) ˜Error bounds for convolutional codes and an asymptotically optimum
decoding algorithm™, IEEE Transactions on Information Theory IT-13, 260“269.
H. Holma, A. Toskala (2002) WCDMA for UMTS, 2nd edn. John Wiley&Sons, Chichester.
J. Laiho, A. Wacker, T. Novosad (2002) Radio Network Planning and Optimisation for
UMTS, John Wiley&Sons, Chichester.
A list of the current versions of the speci¬cations can be found at http://www.3gpp.org/
specs/web-table specs-with-titles-and-latest-versions.htm, and the 3GPP ftp site for the
individual speci¬cation documents is http://www.3gpp.org/ftp/Specs/latest/
GSM Fundamentals

The roots of the development of the global system for mobile communications (GSM)
began with a group formed by the European Conference of Postal and Telecommuni-
cations Administrations (CEPT) to investigate the development of a standard mobile
telephone system to be used throughout Europe. This group was known as the Groupe
Special Mobile, or GSM for short, and this is initially where the acronym GSM came
from; however, it now is widely understood to stand for global system for mobile com-
munications. A uni¬ed telephone system was desirable since Europe is made up of many
separate countries each with their own government, language, culture and telecommuni-
cation infrastructure, much of which was still in the hands of state-run monopolies. As
there is much trade between these countries, a mobile network which would free users to
roam internationally from country to country was seen as a valuable asset.
The other major region to discuss in parallel is movements in mobile communications
in the USA. Mobile technology was advancing there also, but the motivation to provide
roaming capabilities was not such a fundamental requirement, since it is one country.
There was and is considerable regionalization of communications in the USA and this
was re¬‚ected in the proliferation of mobile devices, where operators only needed to cater
for the domestic market.
GSM was eventually adopted as a European standard by the European Telecommuni-
cations Standards Institute (ETSI). It has been standardized to operate on three principal
frequency regions, being 900 MHz, 1800 MHz and 1900 MHz.
GSM is by far the most successful of the second generation cellular systems, and has
seen widespread adoption not only across Europe but also throughout the Asia-Paci¬c
region, and more recently, the Americas. Some of the large mobile network operators
in the USA are also introducing GSM, either as a migration step towards the UMTS
¬‚avour of 3G or simply in addition to the current offerings. Two such operators that
are deploying GSM are Voicestream and AT&T wireless. Other existing systems include
IS-136 (TDMA), IS-95 or cdmaOne (CDMA), and PDC. Japan tried to mimic the success
of GSM with its home-grown PDC system, intending for this to spread throughout Asia.

Convergence Technologies for 3G Networks: IP, UMTS, EGPRS and ATM J. Bannister, P. Mather and S. Coope
™ 2004 John Wiley & Sons, Ltd ISBN: 0-470-86091-X

However, the popularity of GSM brought to bear the economies of scale and thus PDC
is only evident in Japan. The success of GSM in Asia is not surprising as many of the
original ideas in the design of a network that would transcend political borders are also
relevant in Asia.
Currently, at time of writing, GSM technology has over 70% global market share of
second generation cellular systems. As networks evolve to 3G, GSM should not be seen
as becoming redundant, but rather GSM is an integral part of the 3G UMTS network
infrastructure as it also evolves to the GSM-EDGE radio access network (GERAN).

Figure 3.1 shows the general architecture for a GSM network. The various functional
blocks are explained in the following subsections.

Mobile station (MS)
The MS consists of the mobile equipment (ME; the actual device) and a smart card called
the subscriber identity module (SIM). The SIM offers personal mobility since the user
can remove the SIM card from one mobile device and place it in another device without
informing the network operator. In contrast, most other 2G systems require a registration
update to the operator. The SIM contains a globally unique identi¬er, the international
mobile subscriber identity (IMSI), as well as a secret key used for authentication and other
security procedures. The IMSI (or a variation of it for security purposes) is used throughout
the network as the identi¬er for the subscriber. This system enables a subscriber to
change the mobile equipment and still be able to make calls, receive calls and receive
other subscriber information by simply transferring the SIM card to the new device.
Any calls made will appear on a single user bill irrespective of changes in the mobile
device. The mobile equipment is also uniquely identi¬able by the international mobile
equipment identity (IMEI). The IMEI and IMSI are independent, thus providing the user
¬‚exibility by separating the concept of subscriber from access device. Many operators still
issue ˜locked™ mobile devices where the equipment is tied for use only on a particular
operator™s network. A mobile device not equipped with a SIM must also still be able to

Mobile PSTN

Figure 3.1 GSM general architecture

IMEI number


Figure 3.2 GSM IMEI and IMSI

make emergency calls. To protect the call from undesirable snooping or listening in, the
IMSI will not always be transmitted over the cell to identify the subscriber. Instead a
temporary IMSI (T-IMSI) identi¬er is used and changed at regular intervals. Note that
for extra security the whole data stream is encrypted over the air interface. Figure 3.2
shows the 15-digit IMEI number on the left and the SIM card, which incorporates the
15-digit IMSI.

Base station subsystem (BSS)
The base station subsystem (BSS) is composed of three parts, the base transceiver station
(BTS), the base station controller (BSC), which controls the BTSs, and the transcoding
and rate adaption unit (TRAU).

Base transceiver station (BTS)
The BTS houses the radio transceivers (TRXs) that de¬ne a cell and handle the radio link
with the mobile station. As was seen, each TRX can handle up to eight full-rate users
simultaneously. If more than eight full-rate users request resources within the TRX then
they will receive a busy tone, or a network busy message may be displayed on the mobile
device. It is possible to increase the number of simultaneous users in a cell by increasing
the number of TRXs, hence the number of frequencies used. When a mobile device moves
from one cell to another the BTS may change. Within the GSM system a mobile device
is connected to only one BTS at a given time. The ¬rst TRX in a cell can actually only
handle a maximum of seven (possibly less) simultaneous users since one channel on the
downlink is used for broadcasting general system information through what is known as
the broadcast and control channel (BCCH). The BTS is also responsible for encrypting
the radio link to the mobile device based on security information it receives from the
core network.

Base station controller (BSC)
The BSC manages the radio resources for one or more BTSs. It handles the radio channel
setup, frequency hopping and handover procedures when a user moves from one cell to
another. When a handover occurs, the BSC may change; it is a design consideration that

this will not change with the same regularity as a BTS change. A BSC communicates
with the BTS through time division multiplex (TDM) channels over what is referred to
as the Abis interface, generally implemented using E1 or T1 lines. If the numerous BTSs
and the corresponding BSC are in close proximity then this link may be a ¬bre optic
or copper cable connection. In some cases, there are a large number of BTSs in close
proximity but quite some distance away from the controlling BSC. In such cases it may
be more ef¬cient to relay the calls from each of the BTSs to a single BTS via microwave
links. This type of link may be very cost effective since generally the running costs of
a point-to-point microwave link may be free. Of course this has to be weighed against
the cost of the purchasing and deployment of the equipment. The collector BTS can then
connect to the BSC via another microwave link or via a landline cable. A problem with
the above system is that if the collector BTS fails then calls from the other BTSs may
also fail. To overcome this problem it is possible to have two collector BTSs both sending
the calls to the BSC. This forms a redundant link and if one collector BTS fails then this
does not present such a large problem, as is illustrated in Figure 3.3(b).

Transcoding and rate adaption unit (TRAU)
The central role of the second generation systems is to transfer speech calls and the
system has been designed and optimized for voice traf¬c. The human voice is converted
to binary in a rather complex process. GSM is now quite an old system and as such
the original encoding method used (LPC-RPE1 ) is not as ef¬cient as some of the more
recently developed coding systems such as those used in other cellular systems. There
have been many developments in digital signal processing (DSP) which have enabled
good voice quality to be transmitted at lower data rates. Although the TRAU is actually

BTS Base
BTS Base
BTS Station BTS


(a) (b)

Figure 3.3 Base station connectivity

Linear predictive coding with regular pulse excitation (LPC-RPE) provides a digital model of the
vocal tract and vocal chords, excited by a signal which is air from the lungs.

seen as being logically part of the BSS, it usually resides close to the MSC since this
has signi¬cant impact on reducing the transmission costs. The voice data is sent in a
16 kbps channel through to the TRAU from the mobile device via the BTS and BSC.
The TRAU will convert this speech to the standard 64 kbps for transfer over the PSTN
or ISDN network. This process is illustrated in Figure 3.4, where over the air interface,
speech uses 13 kbps (full-rate) and data 9.6 or 14.4 kbps, with each of these requiring a
16 kbps link through the BSS.
As has been mentioned, digital voice data is robust in the face of errors, and can han-
dle substantial bit error rates before the user begins to notice signal degradation. This
is in stark contrast to data such as IP packets, which is extremely error intolerant and
a checksum is generally used to drop a packet which contains an error. Table 3.1 lists
the adaptive multirate (AMR) speech CODECS which are implemented in UMTS. Also
indicated on the diagram are the enhanced full-rate (EFR) bit rates for the second gener-
ation GSM, TDMA and PDC systems for comparison. The GSM EFR uses the algebraic
code excited linear prediction (ACELP) algorithm and gives better quality speech than
full-rate (FR) using 12.2 kbps. A half-rate (HR) method of speech coding has also been
introduced in to the standards, which is known as code excited linear prediction-vector
sum excited linear prediction (CELP-VSELP). This method will enable two subscribers
to share a single time slot.

Network switching subsystem (NSS)
The NSS comprises the circuit switched core network part of the GSM system. The main
element is the mobile switching centre (MSC) switch and a number of databases referred

Station 9.6, 13,
14.4 kbps
SIM 16kbps 16kbps 64kbps 64kbps

Figure 3.4 Transcoding

Table 3.1 CODEC bit rates
CODEC Bit rate (kbps)
AMR 12.20 12.2 (GSM EFR)
AMR 10.20 10.2
AMR 7.95 7.95
AMR 7.40 7.4 (TDMA EFR)
AMR 6.70 6.7 (PDC EFR)
AMR 5.90 5.9
AMR 5.15 5.15
AMR 4.75 4.75

to as the visitor location register (VLR) and home location register (HLR). The HLR is
always in the home network for roaming subscribers and thus any data exchange may
have to cross international boundaries. The MSC and VLR are usually combined and are
located in the visited network.

Mobile switching centre (MSC)
This acts like a normal switching node for a PSTN or ISDN network. It also takes care
of all the additional functionality required to support a mobile subscriber. It therefore has
the dual role of both switching and management. When a mobile device is switched on
and requests a connection to a mobile network, it is principally the MSC that processes
this request, with the BSS merely providing the access to facilitate this request. If the
request is successful then the MSC registers the mobile device within its associated VLR
(see below; most manufacturers tend to combine the VLR functionality with the MSC).
The VLR will update the HLR with the location of this mobile device, and the HLR
may be either in the same network, or a different network in the case of a roaming user.


. 11
( 132 .)