The
development of GSM started in the early 1980s. It was seen then as the mainstay
of the plans for Europe´s mobile communication infrastructure for the 1990s.
Today, GSM and its DCS 1800 and PCS 1900 versions have spread far beyond
Western Europe with networks installed across all continents.
The story begins in 1982
when the European Conference of Posts and Telecommunications Administrations
(CEPT), consisting then of the telecommunication administrations of
twenty six nations made two very significant decisions. The first was
to establish a team with the title "Groupe Spéciale Mobile"
(hence the term "GSM", which today stands for Global System
for Mobile Communications) to develop a set of common standards for
a future pan-European cellular network. The second was to recommend
that two blocks of frequencies in the 900 MHz band be set aside for
the system. The CEPT made these decisions
in an attempt to solve the problems created by the uncoordinated development
of individual national mobile communication systems using incompatible
standards. The impossibility of using the same terminal in different
countries whilst traveling across Europe was one of these problems;
another was the difficulty of establishing a Europe-wide mobile communications
industry that would be competitive in world markets due to the lack
of a sufficiently larger home market with common standards - with its
attendant economies of scale. By 1986 it was clear that
some of these analogue cellular networks would run out of capacity by
the early 1990s. As a result, a directive was issued for two blocks
of frequencies in the 900 MHz band, albeit somewhat smaller than recommended
by the CEPT, to be reserved absolutely for a pan-European service to
be opened in 1991. In the meantime the GSM members
were making excellent progress with the development of agreed standards.
One major decision was to adopt a digital rather than an analogue system.
The digital system would offer improved spectrum efficiency, better
quality transmission and new services with enhanced features including
security. It would also permit the use of Very Large Scale Integration
(VLSI) technology which would lead to smaller and cheaper mobiles, including
hand held terminals. Finally, a digital approach would complement the
development of the Integrated Services Digital Network (ISDN) with which
GSM would have to interface. GSM initially stood
for Group Spécial Mobile, the CEPT (Conference of European Posts &
Telegraphs) formed the group to develop a Pan-European cellular system
to replace the many systems already in place in Europe that were all
incompatible. The main features
of GSM were to be International Roaming ability, good sound quality,
small cheap handsets and ability to handle high volumes of users. GSM
was taken over in 1989 by the ETSI (European Telecommunications Standards
Institute) and they finalised the GSM standard in 1990. GSM service
started in 1991. It was also renamed this year to Global System for
Mobile communications (GSM). Today there are
approx. 105 countries with GSM networks or planned networks and many
more are planned with around 32 million subscribers world wide on the
139 networks. This accounts for over 25% of the world's cellular market. The MoU "Memorandum
of Understanding" has over 210 members from 105 countries, this
organisation meets ever three to four months to look at new or better
implementations to the GSM system. The MoU has a website
that goes into more details at http://www.gsmworld.com.
The most important
events in the development of the GSM system are presented in the table
1. From the evolution
of GSM, it is clear that GSM is not anymore only a European standard. GSM
networks are operationnal or planned in over 80 countries around the world.
The rapid and increasing acceptance of the GSM system is illustrated with
the following figures:
In a cellular
system, the covering area of an operator is divided into cells. A cell
corresponds to the covering area of one transmitter or a small collection
of transmitters. The size of a cell is determined by the transmitter's
power.
The concept of
cellular systems is the use of low power transmitters in order to enable
the efficient reuse of the frequencies. In fact, if the transmitters used
are very powerful, the frequencies can not be reused for hundred of kilometers
as they are limited to the covering area of the transmitter.
The frequency
band allocated to a cellular mobile radio system is distributed over a
group of cells and this distribution is repeated in all the covering area
of an operator. The whole number of radio channels available can then be
used in each group of cells that form the covering area of an operator.
Frequencies used in a cell will be reused several cells away. The distance
between the cells using the same frequency must be sufficient to avoid
interference. The frequency reuse will increase considerably the capacity
in number of users.
In order to work
properly, a cellular system must verify the following two main conditions:
.
Year
Events
1982
CEPT establishes a GSM group in order to develop the standards
for a pan-European cellular mobile system
1985
Adoption of a list of recommendations to be generated by
the group
1986
Field tests were performed in order to test the different
radio techniques proposed for the air interface
1987
TDMA is chosen as access method (in fact, it will be used
with FDMA) Initial Memorandum of Understanding (MoU) signed by telecommunication
operators (representing 12 countries)
1988
Validation of the GSM system
1989
The responsability of the GSM specifications is passed
to the ETSI
1990
Appearance of the phase 1 of the GSM specifications
1991
Commercial launch of the GSM service
1992
Enlargement of the countries that signed the GSM- MoU>
Coverage of larger cities/airports
1993
Coverage of main roads GSM services start outside Europe
1995
Phase 2 of the GSM specifications Coverage of rural areas
Since the appearance
of GSM, other digital mobile systems have been developed. The table 2 charts
the different mobile cellular systems developed since the commercial launch
of cellular systems.
Year
Mobile Cellular System
1981
Nordic Mobile Telephony (NMT), 450>
1983
American Mobile Phone System (AMPS)
1985
Total Access Communication System (TACS) Radiocom 2000
C-Netz
1986
Nordic Mobile Telephony (NMT), 900>
1991
Global System for Mobile communications> North American
Digital Cellular (NADC)
1992
Digital Cellular System (DCS) 1800
1994
Personal Digital Cellular (PDC) or Japanese Digital Cellular
(JDC)
1995
Personal Communications Systems (PCS) 1900- Canada>
1996
PCS-United States of America>
GSM: A Cellular System
In order to exchange
the information needed to maintain the communication links within the cellular
network, several radio channels are reserved for the signaling information.
Cluster
The macrocells are large cells for remote and sparsely populated areas.
Microcells
These cells are used for densely populated areas. By splitting the existing areas into smaller cells, the number of channels available is increased as well as the capacity of the cells. The power level of the transmitters used in these cells is then decreased, reducing the possibility of interference between neighboring cells.
Selective
cells
It is not always
useful to define a cell with a full coverage of 360 degrees. In some cases,
cells with a particular shape and coverage are needed. These cells are
called selective cells. A typical example of selective cells are the cells
that may be located at the entrances of tunnels where a coverage of 360
degrees is not needed. In this case, a selective cell with a coverage of
120 degrees is used.
Umbrella
cells
A freeway crossing
very small cells produces an important number of handovers among the different
small neighboring cells. In order to solve this problem, the concept of
umbrella cells is introduced. An umbrella cell covers several microcells.
The power level inside an umbrella cell is increased comparing to the power
levels used in the microcells that form the umbrella cell. When the speed
of the mobile is too high, the mobile is handed off to the umbrella cell.
The mobile will then stay longer in the same cell (in this case the umbrella
cell). This will reduce the number of handovers and the work of the network.
The GSM technical
specifications define the different entities that form the GSM network
by defining their functions and interface requirements.
The GSM network
can be divided into four main parts:
GSM: System Architecture
The architecture of the GSM network
is presented in figure 1.
Mobile Station (MS)
A Mobile Station consists of two main elements:
The Terminal
There are different types of terminals distinguished principally by their power and application:
 |
If you take a cell phone apart, you find that it contains just a few individual parts:
 The front of the circuit board |
 The back of the circuit board |
In the photos above, you see several computer chips. Let's talk about what some of the individual chips do. The analog-to-digital and digital-to-analog conversion chips translate the outgoing audio signal from analog to digital and the incoming signal from digital back to analog. The digital signal processor (DSP) is a highly customized processor designed to perform signal-manipulation calculations at high speed.
 The microprocessor |
The microprocessor handles all of the housekeeping chores for the keyboard and display, deals with command and control signaling with the base station and also coordinates the rest of the functions on the board. The ROM and Flash memory chips provide storage for the phone's operating system and customizable features, such as the phone directory. The radio frequency (RF) and power section handles power management and recharging, and also deals with the hundreds of FM channels. Finally, the RF amplifiers handle signals traveling to and from the antenna.
 The display and keypad contacts |
The display has grown considerably in size as the number of features in cell phones have increased. Most current phones offer built-in phone directories, calculators and even games. And many of the phones incorporate some type of PDA or Web browser.
 The Flash memory card on the circuit board |
 The Flash memory card removed |
Some phones store certain information, such as the SID and MIN codes, in internal Flash memory, while others use external cards that are similar to SmartMedia cards.
 The cell-phone speaker, microphone and battery backup |
Cell phones have such tiny speakers and microphones that it is incredible how well most of them reproduce sound. As you can see in the picture above, the speaker is about the size of a dime and the microphone is no larger than the watch battery beside it. Speaking of the watch battery, this is used by the cell phone's internal clock chip.
What is amazing is that all of that functionality -- which only 30 years ago would have filled an entire floor of an office building -- now fits into a package that sits comfortably in the palm of your hand!
The SIM
The SIM is a smart card that identifies the terminal. By inserting the SIM card into the terminal, the user can have access to all the subscribed services. Without the SIM card, the terminal is not operational.
The SIM card is protected by a four-digit Personal Identification Number (PIN). In order to identify the subscriber to the system, the SIM card contains some parameters of the user such as its International Mobile Subscriber Identity (IMSI).
Another advantage of the SIM card is the mobility of the users. In fact, the only element that personalizes a terminal is the SIM card. Therefore, the user can have access to its subscribed services in any terminal using its SIM card.
Handshake between SIM and Base Station
Language data KC |
||
|
|
|
SIM File Structure
SIM has hierarchically developed file system. It has the Master File (MF) and two Dedicated Files (DF), in which the Elementary Filess (EF) with the data for applications are stored.
SIM File Identifier
Dedicated Files (DF)
A Dedicated File (DF) is a functional grouping of files consisting of itself and all those files which contain this DF in their parental hierarchy (that is to say it consists of the DF and its complete "subtree"). A DF "consists" only of a header part. Two DFs are defined in this specification:Elementary Files (EF)
An Elementary File (EF) is composed of a header and a body part. The following three structures of an EF are used by GSM:
File Identifier and Directory Structure inside GSM SIM
Base Station Subsystem (BSS)
The BSS connects the Mobile Station and the NSS. It is in charge of the transmission and reception. The BSS can be divided into two parts:
The Base Transceiver Station houses the radio tranceivers that define a cell and handles the radio-link protocols with the Mobile Station. In a large urban area, there will potentially be a large number of BTSs deployed, thus the requirements for a BTS are ruggedness, reliability, portability, and minimum cost. BTS are all connected together to allow you to move from one cell to another. A BTS is usually placed in the center of a cell. Its transmitting power defines the size of a cell. Each BTS has between one and sixteen transceivers depending on the density of users in the cell.
The Cell Site Antenna
"Depending on the best signal strength received, usually, the nearest cell site antenna makes a radio connection with our phone . . ." Down the antenna our signal goes!
Base Satation
Here's where call comes to, after being picked up just a short distance away by the cell site antenna. BS stands for base station. BTS means Base Transceiver Station which is the same thing. RBS for Remote Base station is also used, sometimes for smaller base stations. Whatever the name, this is the radio gear that passes all calls coming in and going out of a cell site. The BTS lies at the heart of the cell site.So, the base station sends out calls to the mobiles and receives the return traffic. It sends and returns those calls to the regular telephone network by way of a two step process. The base station is under direction of a base station controller so traffic gets sent there first. The base station controller, described below, gathers the calls from many base stations and passes them on to a mobile telephone switch. From that switch come and go the calls from the regular telephone network. The figure below shows a Siemans BTS.
The BSC controls a group of BTS and manages their radio ressources. A BSC is principally in charge of handovers, frequency hopping, exchange functions and control of the radio frequency power levels of the BTSs. BSC is a modular, digital switching plateform that connects the Mobile Switching Center (MSC) and Base Tranceiver Stations (BTS). BSC also transfers signalling information to and from Mobile Station (MS) and manages handover between cells. The BSC is the radio network's "brain", maintaining high capacity and quality. The BSC unit also performs transcoding functions to convert between 64Kbps channel rate used in the Switching System and the 16Kbps channel rate for GSM traffic.
If every base station talked directly to the MSC, traffic would become too congested. To ensure quality communications via traffic management, the wireless infrastructure network uses Base Station Controllers as a way to segment the network and control congestion. The result is that MSCs route their circuits to BSCs which in turn are responsible for connectivity and routing of calls for 50 to 100 wireless base stations. The figure below shows a Nokia BSC.
Networking and Switching Subsystem (NSS)
Its main role is to manage the communications between the mobile users and other users, such as mobile users, ISDN users, fixed telephony users, etc. It also includes data bases needed in order to store information about the subscribers and to manage their mobility. The different components of the NSS are described below.
It is the central component of the NSS. The MSC performs the switching functions that connect mobile subscribers to fixed network subscribers or to other mobile subscribers. The MSC is connected to fixed networks such as the PSTN, ISDN, etc. The MSC retrieves the data needed to process subscriber requests from different types of databases, including the Visitor Location Register (VLR), Home Location Register (HLR), and Authentication Center (AuC).
The MSC generates all billing records and ensures that all usage is directed to the appropriate account. The MSC has a relatively complex task, as unlike a conventional telephone exchange, when GSM subscribers make calls they could be anywhere within the network. The MSC must ensure that calls are routed through to those subscribers, wherever they are and wherever they move to throughout the duration of each cell. This situation becomes even more complex when two mobile subscribers wish to contact each other from two distant locations. In order to simplify the subscriber management function, a specific service area is allocated to each MSC. The MSC has to control the switching of tariff to and from the subscribers within it's service area which involves the coordination of all radio resources and the inter cell hand-off activities. The figure below shows Nokia MSC's.
The Home Location Register (HLR) is the main database of permanent subscriber information for a mobile network. The HLR is an integral component of CDMA (code division multiple access), TDMA (time division multiple access), and GSM (Global System for Mobile communications) networks. Maintained by the subscriber's home carrier (or the network operator where the user initiated the call), the HLR contains pertinent user information, including address, account status, and preferences. The HLR interacts with the Mobile Switching Center (MSC), which is a switch used for call control and processing.
HLR offers flexible dimensioning, telecom grade performance and profit generating features. HLR is designed to be a flexible application enabling features in the network to meet the demand of existing and future converging networks. HLR provides realtime performance based on a well proven platform. HLR manages mobile subscriber profiles as well as subscriber location and activity, and also handles supplementary services.
HLR serves one or more Mobile Switching Centers (MSC). HLR can be collocated with MSC and with other applications,like Authentication Center (AUC). The flexibility lets operators optimize the network,depending on present and future needs, making it a very cost-effective solution. HLR can optionally be configured for geographical redundancy,minimizing the risk of loss during natural or human-caused disasters.
The two key references used to route calls to each subscriber are the International Mobile Subscriber Identity (IMSI) and the Mobile Subscriber Integrated Services Digital Network (MSISDN) number. The IMSI is the unique number allocated to the subscriber which is stored in the SIM Card and is used by the network for internal communications. When the SIM Card is inserted into a Mobile Equipment it becomes a Mobile Station. The MSISDN is the subscriber's mobile number which is linked to the IMSI in the HLR. Incoming calls to a subscriber are translated back to the IMSI at the HLR thus enabling them to be delivered to the Mobile Station. Once the Mobile Station's MSISDN has been used to identify the IMSI, the HLR verifies the subscription records to ensure that the call can be delivered to the last known location of the Mobile Station.
The HLR is a big database maintained on computers called servers, often UNIX workstations. The figure below shows an Erricson HLR.
Visitors Location Register (VLR)
The VLR is a database that is linked to an MSC and temporarily stores information about each Mobile Station within the area served by that MSC. The information that is temporarily stored in the VLR is sufficient to allow any Mobile Station within that MSC area to make and receive calls. This includes the Mobile Station's identity, the area in which it was last registered and data pertaining to the subscriber and any supplementary services that have been selected by the subscriber. The MSC refers to the VLR each time that a Mobile Station attempts to make a call in order to verify that the request can be fulfilled. This process is to establish that no call restrictions or call barring instructions are in place.
The VLR contains roamer information. Once the visited system detects your mobile, its VLR queries your assigned home location register. The VLR makes sure you are a valid subscriber, then retrieves just enough information from the now distant HLR to manage your call. It temporarily stores your last known location area, the power your mobile uses, special services you subscribe to and so on. Though traveling, the cellular network now knows where you are and can direct calls to you.
The following elements can be found in VLR database tables:
The EIR ensures that all Mobile Equipment's are valid and authorised to function on the PLMN. Three categories exist on the EIR, a white list, a gray list and a black list. The white list comprises the IMEI ranges of all the Mobile Equipment's that have been approved by any one of the three European, GSM approval centers. Any Mobile Equipment that appears on the gray list will be allowed to function but will trigger an alert to the network operator. This facility allows the network operator to identify any subscriber that is using a lost or stolen Mobile Equipment. Mobiles that are lost or stolen can be blacklisted which will prevent them from functioning on the home PLMN or on other PLMNs around the world.
In order to get IMIE of your mobile phone press * # 0 6 # on your mobile phone.
A central EIR is managed by the MoU Permanent Secretariat in Dublin, Ireland. Every MoU member is committed to linking their network's EIR to the CEIR. The advantage in having the CEIR concept is that it empowers each network operator to restrict or prevent the operation of any given MS throughout all PLMNs that are linked up to the CEIR.
The AC or AuC is the Authentication Center, a secured database handling authentication and encryption keys. Authentication verifies a mobile customer with a complex challenge and reply routine. The network sends a randomly generated number to the mobile. The mobile then performs a calculation against it with a number it has stored in its SIM and sends the result back. Only if the switch gets the number it expects does the call proceed. The AC stores all data needed to authenticate a call and to then encrypt both voice traffic and signaling messages. The AUC and EIR are implemented as stand-alone nodes or as a combined AUC/EIR node.
Message Center (MXE)
The MXE is a node that provides integrated voice, fax, and data messaging. Specifically, the MXE handles short message service, cell broadcast, voice mail, fax mail, e-mail, and notification.Mobile Service Node (MSN)
The MSN is the node that handles the mobile intelligent network (IN) services.Gateway Mobile Services Switching Center (GMSC)
A gateway is a node used to interconnect two networks. The gateway is often implemented in an MSC. The MSC is then referred to as the GMSC. The GMSC is the point to which a MS terminating call is initially routed, without any knowledge of the MS's location. The GMSC is thus in charge of obtaining the MSRN (Mobile Station Roaming Number) from the HLR based on the MSISDN (Mobile Station ISDN number, the "directory number" of a MS) and routing the call to the correct visited MSC. The "MSC" part of the term GMSC is misleading, since the gateway operation does not require any linking to a MSC.GSM interworking unit (GIWU)
The GIWU consists of both hardware and software that provides an interface to various networks for data communications. Through the GIWU, users can alternate between speech and data during the same call. The GIWU hardware equipment is physically located at the MSC/VLR.The Operation and Support Subsystem (OSS)
The operations and maintenance center (OMC) is connected to all equipment in the switching system and to the BSC. The implementation of OMC is also known as operation and support system (OSS). The OSS is the functional entity from which the network operator monitors and controls the system. The purpose of OSS is to offer the customer cost-effective support for centralized, regional, and local operational and maintenance activities that are required for a GSM network. An important function of OSS is to provide a network overview and support the maintenance activities of different operation and maintenance organizations. It is also in charge of controlling the traffic load of the BSS.
| Um | The air interface is used for exchanges between a MS and a BSS. LAPDm,
a modified version of the ISDN LAPD, is used for signalling. |
|---|---|
| Abis | This is a BSS internal interface linking the BSC and a BTS, and it
has not been standardised. The Abis interface allows control of the radio
equipment and radio frequency allocation in the BTS. |
| A | The A interface is between the BSS and the MSC. The A interface manages
the allocation of suitable radio resources to the MSs and mobility management. |
| B | The B interface between the MSC and the VLR uses the MAP/B protocol.
Most MSCs are associated with a VLR, making the B interface "internal".
Whenever the MSC needs access to data regarding a MS located in its area,
it interrogates the VLR using the MAP/B protocol over the B interface.
|
| C | The C interface is between the HLR and a GMSC or a SMS-G. Each call
originating outside of GSM (i.e., a MS terminating call from the PSTN)
has to go through a Gateway to obtain the routing information required
to complete the call, and the MAP/C protocol over the C interface is used
for this purpose. Also, the MSC may optionally forward billing information
to the HLR after call clearing. |
| D | The D interface is between the VLR and HLR, and uses the MAP/D protocol
to exchange the data related to the location of the MS and to the management
of the subscriber. |
| E | The E interface interconnects two MSCs. The E interface
exchanges data related to handover between the anchor and relay MSCs using
the MAP/E protocol. |
| F | The F interface connects the MSC to the EIR, and uses the MAP/F protocol
to verify the status of the IMEI that the MSC has retrieved from the MS. |
| G | The G interface interconnects two VLRs of different MSCs and uses the
MAP/G protocol to transfer subscriber information, during e.g. a location
update procedure. |
| H | The H interface is between the MSC and the SMS-G, and uses the MAP/H
protocol to support the transfer of short messages. |
| I | The I interface (not shown in the Figure ) is the interface
between the MSC and the MS. Messages exchanged over the I interface are
relayed transparently through the BSS. |
For the GSM-900 system, two frequency bands have been made available:
The 25 MHz bands are then divided into 124 pairs of frequency duplex channels with 200 kHz carrier spacing using Frequency Division Multiple Access (FDMA). Since it is not possible for a same cell to use two adjacent channels, the channel spacing can be said to be 200 kHz interleaved. One or more carrier frequencies are assigned to individual Base Station (BS) and a technique known as Time Division Multiple Access (TDMA) is used to split this 200 kHz radio channel into 8 time slots (which creates 8 logical channels). A logical channel is therefore defined by its frequency and the TDMA frame time slot number. By employing eight time slots, each channel transmits the digitized speech in a series of short bursts: a GSM terminal is only ever transmitting for one eighth of the time.
8-slot TDMA together with the 248 physical half-duplex channels corresponds to a total of 1984 logical half-duplex channels. This corresponds to roughly 283 (1984 / 7) logical half-duplex channels per cell. This is because a cell can only use one seventh of the total number of frequencies, see Figure below.
Figure : Typical cellular scheme
Seven sets of frequencies are sufficient to cover an arbitrarily large area, providing that the repeat-distance d is larger than twice the maximum radius r covered by each transmitter.
Each of the frequency channels is segmented into 8 time slots of length 0.577 ms (15/26 ms). The 8 time slots makes up a TDMA frame of length 4.615 ms (120/26 ms). The recurrence of one particular time slot every 4.615 ms makes up one basic channel.
The GSM system distinguishes between traffic channels (used for user data) and control channels (reserved for network management messages). The Traffic Channel/Full-Rate Speech (TCH/FS) is used to carry speech at 13 kbps.
TCHs for the uplink and downlink are separated in time by 3 burst periods, so that the mobile does not has to transmit and receive simultaneously. TCHs are defined using a 26-frame multiframe (i.e. a group of 26 TDMA frames). The length of a 26-frame multiframe is 120 ms, which is how the length of a burst period is defined (120 ms / 26 frames / 8 burst periods per frame). Out of the 26 frames, 24 are used for traffic, one is used for the Slow Associated Control Channel (SACCH) and one is currently unused.
Figure The TDMA frame structure
Data are transmitted in bursts which are placed within the time slots. The transmission bit rate is 271 kb/s (bit period 3.79 microseconds). To allow for time alignment errors, time dispersion etc, the data burst is slightly shorter than the time slot (148 out of the 156.25 bit periods available within a time slot).
The burst is the transmission quantum of GSM. Its transmission takes place during a time window lasting (576 + 12/13) microseconds, i.e. (156 + 1/4) bit duration. A normal burst contains two packets of 58 bits (57 data bits + 1 stealing bit) surrounding a training sequence of 26 bits. The 26-bit training sequence is of a known pattern that is compared with the received pattern in order to reconstruct the rest of the original signal (multipath equalization). The actual implementation of the equalizer is not specified in the GSM specifications. Three ``tail'' bits are added on each side.
GSM can use slow frequency hopping where the mobile station and the base station transmit each TDMA frame on a different carrier frequency. The frequency hopping algorithm is broadcast on the Broadcast Control Channel. Since multipath fading is dependent on carrier frequency, slow frequency hopping help mitigate the problem. Frequency hopping is an option for each individual cell and a base station is not required to support this feature.
Common channels can be accessed both by idle mode and dedicated mode mobiles. The common channels are used by idle mode mobiles to exchange the signalling information required to change to dedicated mode. Mobiles already in dedicated mode monitor the surrounding base stations for handover and other information. The common channels are defined within a 51-frame multiframe, so that dedicated mobiles using the 26-frame multiframe TCH structure can still monitor control channels. The common channels include:
Speech to Radio Conversion
Figure below depicted the sequence of operations from speech to radio waves and from radio waves to speech. These operations are described in the following sections.
Figure : The sequence of operations
GSM Speech Coding
The full rate speech codec in GSM is described as Regular Pulse Excitation with Long Term Prediction (GSM 06.10 RPE-LTP). Basically, the encoder divides the speech into short-term predictable parts, long-term predictable part and the remaining residual pulse. Then, it encodes that pulse and parameters for the two predictors. The decoder reconstructs the speech by passing the residual pulse first through the long-term prediction filter, and then through the short-term predictor, see Figure below.
Figure : A block diagram of the GSM 06.10 codec
GSM is a digital system, so speech which is inherently analog, has to be digitized. The method employed by ISDN, and by current telephone systems for multiplexing voice lines over high speed trunks and optical fiber lines, is Pulse Coded Modulation (PCM). The output stream from PCM is 64 kbps, too high a rate to be feasible over a radio link. The 64 kbps signal, although simple to implement, contains much redundancy. The GSM group studied several speech coding algorithms on the basis of subjective speech quality and complexity (which is related to cost, processing delay, and power consumption once implemented) before arriving at the choice of a Regular Pulse Excited -- Linear Predictive Coder (RPE--LPC) with a Long Term Predictor loop. Basically, information from previous samples, which does not change very quickly, is used to predict the current sample. The coefficients of the linear combination of the previous samples, plus an encoded form of the residual, the difference between the predicted and actual sample, represent the signal. Speech is divided into 20 millisecond samples, each of which is encoded as 260 bits, giving a total bit rate of 13 kbps. This is the so-called Full-Rate speech coding. Recently, an Enhanced Full-Rate (EFR) speech coding algorithm has been implemented by some North American GSM1900 operators. This is said to provide improved speech quality using the existing 13 kbps bit rate.
GSM Channel Coding
Because of natural and man-made electromagnetic interference, the encoded speech or data signal transmitted over the radio interface must be protected from errors. Channel coding introduces redundancy into the data flow in order to allow the detection or even the correction of bit errors introduced during the transmission. GSM uses convolutional encoding and block interleaving to achieve this protection. The exact algorithms used differ for speech and for different data rates. The method used for speech blocks will be described below.
The speech coding algorithm produces a speech block of 260 bits every 20 ms (i.e. bit rate 13 kbit/s). In the decoder, these speech blocks are decoded and converted to 13 bit uniformly coded speech samples. The 260 bits of the speech block are classified into two groups. The 78 Class II bits are considered of less importance and are unprotected. The 182 Class I bits are split into 50 Class Ia bits and 132 Class Ib bits (See Figure below).
Figure : Audio sample: 1 block = 260 bits (20 ms)
Class Ia bits are first protected by 3 parity bits for error detection. Class Ib bits are then added together with 4 tail bits before applying the convolutional code with rate and constraint length K=5. The resulting 378 bits are then added to the 78 unprotected Class II bits resulting in a complete coded speech frame of 456 bits (see Figure below).
Figure : TCH/FS Transmission Mode
GSM Error Detecting Codes
The GSM standard uses a 3-bit error redundancy code to enable assessment of the correctness of the bits which are more sensitive to errors in the speech frame (the category Ia 50-bits). If one of these bits are wrong, this may create a loud noise instead of the 20 ms speech slice. Detecting such errors allows the corrupted block to be replaced by something less disturbing (such as an extrapolation of the preceding block).
The polynomial representing the detection code for category Ia bits is
.
At the receiving side, the same operation is done and if the remainder differs, an error is detected and the audio frame is eventually discarded.
GSM Convolutional Coding / Decoding
Convolutional coding consists in transmitting the results of convolutions
of the source sequence using different convolution formulas. The GSM
convolutional code consists in adding 4 bits (set to ``0'') to the initial
185 bit sequence and then applying two different convolutions:
polynomials are respectively 
and
.
The final result is composed of a couple of 189 bits sequences.
Convolutional decoding can be performed using a Viterbi algorithm.
A Viterbi decoder logically explores in parallel every possible user data in
sequence. It encodes and compare each one against the received sequence and
picks up the closest match: it is a maximum likelihood decoder. To reduce the
complexity (the number of possible data sequence double with each additional
data bit), the decoder recognizes at each point that certain sequences cannot
belong to the maximum likelihood path and it discards them.
The encoder memory is limited to K bits; a Viterbi decoder in steady-state
operation keeps only 
paths. Its complexity increases exponentially
with the constraint length K.
The GSM convolutional coding rate per data flow is 378 bits each 20 ms, i.e.: 18.9 kb/s. However, before modulate this signal, the 78 unprotected Class II bits are added. So, the GSM bit rate per flow is 456 bits each 20 ms i.e. 22.8 kb/s.
Interleaving / De-interleaving
Interleaving is meant to decorrelate the relative positions of the bits respectively in the code words and in the modulated radio bursts. The aim of the interleaving algorithm is to avoid the risk of loosing consecutive data bits. GSM blocks of full rate speech are interleaved on 8 bursts: the 456 bits of one block are split into 8 bursts in sub-blocks of 57 bits each. A sub-block is defined as either the odd- or the even-numbered bits of the coded data within one burst. Each sub-blocks of 57 bit is carried by a different burst and in a different TDMA frame. So, a burst contains the contribution of two successive speech blocks A and B. In order to destroy the proximity relations between successive bits, bits of block A use the even positions inside the burst and bits of block B, the odd positions (see Figure below).
Figure : Interleaving operation
De-interleaving consists in performing the reverse operation. The major drawback of interleaving is the corresponding delay: transmission time from the first burst to the last one in a block is equal to 8 TDMA frames (i.e. about 37 ms).
Ciphering / Deciphering
A protection has been introduced in GSM by means of transmission ciphering. The ciphering method does not depend on the type of data to be transmitted (speech, user data or signaling) but is only applied to normal bursts.
Ciphering is achieved by performing an ``exclusive or'' operation between a pseudo-random bit sequence and 114 useful bits of a normal burst (i.e. all information bits except the 2 stealing flags). The pseudo-random sequence is derived from the burst number and a key session established previously through signaling means. Deciphering follows exactly the same operation.
Modulation / De-modulation
GSM uses the Gaussian Minimum Shift Keying (GMSK) with modulation index
, BT (filter bandwidth times bit period) equal to 0.3
and a modulation rate of 271 (270 5/6) kbauds.
The GMSK modulation has been chosen as a compromise between a fairly high
spectrum efficiency (of the order of 1 bit/Hz) and a reasonable demodulation
complexity. The constant envelope allows the use of simple power amplifiers
and the low out-of-band radiation minimizes the effect of adjacent channel
interference. GMSK differs from Minimum Shift Keying (MSK) in that a
pre-modulation Gaussian filter is used. The time-domain impulse response of the
filter is described in Equation 1, where 
and B is the half-power bandwidth.
A block diagram of a GMSK modulator in Figure below.
Figure : GMSK modulation block diagram
