FDMA and TDMA

Using FDMA, a frequency is assigned to a user. So the larger the number of users in a FDMA system, the larger the number of available frequencies must be. The limited available radio spectrum and the fact that a user will not free its assigned frequency until he does not need it anymore, explain why the number of users in a FDMA system can be "quickly" limited.

On the other hand, TDMA allows several users to share the same channel. Each of the users, sharing the common channel, are assigned their own burst within a group of bursts called a frame. Usually TDMA is used with a FDMA structure.

In GSM, a 25 Mhz frequency band is divided, using a FDMA scheme, into 124 carrier frequencies spaced one from each other by a 200 khz frequency band. Normally a 25 Mhz frequency band can provide 125 carrier frequencies but the first carrier frequency is used as a guard band between GSM and other services working on lower frequencies. Each carrier frequency is then divided in time using a TDMA scheme. This scheme splits the radio channel, with a width of 200 khz, into 8 bursts. A burst is the unit of time in a TDMA system, and it lasts approximately 0.577 ms. A TDMA frame is formed with 8 bursts and lasts, consequently, 4.615 ms. Each of the eight bursts, that form a TDMA frame, are then assigned to a single user.

 
   Channel structure

A channel corresponds to the recurrence of one burst every frame. It is defined by its frequency and the position of its corresponding burst within a TDMA frame. In GSM there are two types of channels:

• The traffic channels used to transport speech and data information.
• The control channels used for network management messages and some channel maintenance tasks.

 

      Traffic channels (TCH)

Full-rate traffic channels (TCH/F) are defined using a group of 26 TDMA frames called a 26-Multiframe. The 26-Multiframe lasts consequently 120 ms. In this 26-Multiframe structure, the traffic channels for the downlink and uplink are separated by 3 bursts. As a consequence, the mobiles will not need to transmit and receive at the same time which simplifies considerably the electronics of the system.

The frames that form the 26-Multiframe structure have different functions:

• 24 frames are reserved to traffic.
• 1 frame is used for the Slow Associated Control Channel (SACCH).
• The last frame is unused. This idle frame allows the mobile station to perform other functions, such as measuring the signal strength of neighboring cells.

 
Half-rate traffic channels (TCH/H), which double the capacity of the system, are also grouped in a 26-Multiframe but the internal structure is different.

 
       Control channels

According to their functions, four different classes of control channels are defined:

• Broadcast channels.
• Common control channels.
• Dedicated control channels.
• Associated control channels.

 

           Broadcast channels (BCH)

The BCH channels are used, by the base station, to provide the mobile station with the sufficient information it needs to synchronize with the network. Three different types of BCHs can be distinguished:

• The Broadcast Control Channel (BCCH), which gives to the mobile station the parameters needed in order to identify and access the network
• The Synchronization Channel (SCH), which gives to the mobile station the training sequence needed in order to demodulate the information transmitted by the base station
• The Frequency-Correction Channel (FCCH), which supplies the mobile station with the frequency reference of the system in order to synchronize it with the network

 

           Common Control Channels (CCCH)

The CCCH channels help to establish the calls from the mobile station or the network. Three different types of CCCH can be defined:

• The Paging Channel (PCH). It is used to alert the mobile station of an incoming cal
• The Random Access Channel (RACH), which is used by the mobile station to request access to the network
• The Access Grant Channel (AGCH). It is used, by the base station, to inform the mobile station about which channel it should use. This channel is the answer of a base station to a RACH from the mobile station

 

           Dedicated Control Channels (DCCH)

The DCCH channels are used for message exchange between several mobiles or a mobile and the network. Two different types of DCCH can be defined:

• The Standalone Dedicated Control Channel (SDCCH), which is used in order to exchange signaling information in the downlink and uplink directions.
• The Slow Associated Control Channel (SACCH). It is used for channel maintenance and channel control.

 

           Associated Control Channels

The Fast Associated Control Channels (FACCH) replace all or part of a traffic channel when urgent signaling information must be transmitted. The FACCH channels carry the same information as the SDCCH channels.

 

   Burst structure

As it has been stated before, the burst is the unit in time of a TDMA system. Four different types of bursts can be distinguished in GSM:

• The frequency-correction burst is used on the FCCH. It has the same length as the normal burst but a different structure.
• The synchronization burst is used on the SCH. It has the same length as the normal burst but a different structure.
• The random access burst is used on the RACH and is shorter than the normal burst.
• The normal burst is used to carry speech or data information. It lasts approximately 0.577 ms and has a length of 156.25 bits. Its structure is presented in figure 3.

    figure 3*: Structure of the 26-Multiframe, the TDMA frame and the normal burst

 
The tail bits (T) are a group of three bits set to zero and placed at the beginning and the end of a burst. They are used to cover the periods of ramping up and down of the mobile's power.

The coded data bits corresponds to two groups, of 57 bits each, containing signaling or user data.

The stealing flags (S) indicate, to the receiver, whether the information carried by a burst corresponds to traffic or signaling data.

The training sequence has a length of 26 bits. It is used to synchronize the receiver with the incoming information, avoiding then the negative effects produced by a multipath propagation.

The guard period (GP), with a length of 8.25 bits, is used to avoid a possible overlap of two mobiles during the ramping time.

 

    Frequency hopping

The propagation conditions and therefore the multipath fading depend on the radio frequency. In order to avoid important differences in the quality of the channels, the slow frequency hopping is introduced. The slow frequency hopping changes the frequency with every TDMA frame. A fast frequency hopping changes the frequency many times per frame but it is not used in GSM. The frequency hopping also reduces the effects of co-channel interference.

There are different types of frequency hopping algorithms. The algorithm selected is sent through the Broadcast Control Channels.

Even if frequency hopping can be very useful for the system, a base station does not have to support it necessarily On the other hand, a mobile station has to accept frequency hopping when a base station decides to use it.

 

From source information to radio waves

    The figure 4 presents the different operations that have to be performed in order to pass from the speech source to radio waves and vice versa.
 

 

   
figure 4: From speech source to radio waves  

If the source of information is data and not speech, the speech coding will not be performed.
 

 
   Speech coding

The transmission of speech is, at the moment, the most important service of a mobile cellular system. The GSM speech codec, which will transform the analog signal (voice) into a digital representation, has to meet the following criterias:

• A good speech quality, at least as good as the one obtained with previous cellular systems.
• To reduce the redundancy in the sounds of the voice. This reduction is essential due to the limited capacity of transmission of a radio channel.
• The speech codec must not be very complex because complexity is equivalent to high costs.

 
The final choice for the GSM speech codec is a codec named RPE-LTP (Regular Pulse Excitation Long-Term Prediction). This codec uses the information from previous samples (this information does not change very quickly) in order to predict the current sample. The speech signal is divided into blocks of 20 ms. These blocks are then passed to the speech codec, which has a rate of 13 kbps, in order to obtain blocks of 260 bits.

 

   Channel coding

Channel coding adds redundancy bits to the original information in order to detect and correct, if possible, errors ocurred during the transmission.

 
      Channel coding for the GSM data TCH channels

The channel coding is performed using two codes: a block code and a convolutional code.

The block code corresponds to the block code defined in the GSM Recommendations 05.03. The block code receives an input block of 240 bits and adds four zero tail bits at the end of the input block. The output of the block code is consequently a block of 244 bits.

A convolutional code adds redundancy bits in order to protect the information. A convolutional encoder contains memory. This property differentiates a convolutional code from a block code. A convolutional code can be defined by three variables : n, k and K. The value n corresponds to the number of bits at the output of the encoder, k to the number of bits at the input of the block and K to the memory of the encoder. The ratio, R, of the code is defined as follows : R = k/n. Let's consider a convolutional code with the following values: k is equal to 1, n to 2 and K to 5. This convolutional code uses then a rate of R = 1/2 and a delay of K = 5, which means that it will add a redundant bit for each input bit. The convolutional code uses 5 consecutive bits in order to compute the redundancy bit. As the convolutional code is a 1/2 rate convolutional code, a block of 488 bits is generated. These 488 bits are punctured in order to produce a block of 456 bits. Thirty two bits, obtained as follows, are not transmitted :

    C (11 + 15 j) for j = 0, 1, ..., 31

The block of 456 bits produced by the convolutional code is then passed to the interleaver.

 

       Channel coding for the GSM speech channels

Before applying the channel coding, the 260 bits of a GSM speech frame are divided in three different classes according to their function and importance. The most important class is the class Ia containing 50 bits. Next in importance is the class Ib, which contains 132 bits. The least important is the class II, which contains the remaining 78 bits. The different classes are coded differently. First of all, the class Ia bits are block-coded. Three parity bits, used for error detection, are added to the 50 class Ia bits. The resultant 53 bits are added to the class Ib bits. Four zero bits are added to this block of 185 bits (50+3+132). A convolutional code, with r = 1/2 and K = 5, is then applied, obtaining an output block of 378 bits. The class II bits are added, without any protection, to the output block of the convolutional coder. An output block of 456 bits is finally obtained.

 

       Channel coding for the GSM control channels

In GSM the signalling information is just contained in 184 bits. Forty parity bits, obtained using a fire code, and four zero bits are added to the 184 bits before applying the convolutional code (r = 1/2 and K = 5). The output of the convolutional code is then a block of 456 bits, which does not need to be punctured.

 

   Interleaving

An interleaving rearranges a group of bits in a particular way. It is used in combination with FEC codes in order to improve the performance of the error correction mechanisms. The interleaving decreases the possibility of losing whole bursts during the transmission, by dispersing the errors. Being the errors less concentrated, it is then easier to correct them.

 
       Interleaving for the GSM control channels

A burst in GSM transmits two blocks of 57 data bits each. Therefore the 456 bits corresponding to the output of the channel coder fit into four bursts (4*114 = 456). The 456 bits are divided into eight blocks of 57 bits. The first block of 57 bits contains the bit numbers (0, 8, 16, .....448), the second one the bit numbers (1, 9, 17, .....449), etc. The last block of 57 bits will then contain the bit numbers (7, 15, .....455). The first four blocks of 57 bits are placed in the even-numbered bits of four bursts. The other four blocks of 57 bits are placed in the odd-numbered bits of the same four bursts. Therefore the interleaving depth of the GSM interleaving for control channels is four and a new data block starts every four bursts. The interleaver for control channels is called a block rectangular interleaver.

 
        Interleaving for the GSM speech channels

The block of 456 bits, obtained after the channel coding, is then divided in eight blocks of 57 bits in the same way as it is explained in the previous paragraph. But these eight blocks of 57 bits are distributed differently. The first four blocks of 57 bits are placed in the even-numbered bits of four consecutive bursts. The other four blocks of 57 bits are placed in the odd-numbered bits of the next four bursts. The interleaving depth of the GSM interleaving for speech channels is then eight. A new data block also starts every four bursts. The interleaver for speech channels is called a block diagonal interleaver.

 
       Interleaving for the GSM data TCH channels

A particular interleaving scheme, with an interleaving depth equal to 22, is applied to the block of 456 bits obtained after the channel coding. The block is divided into 16 blocks of 24 bits each, 2 blocks of 18 bits each, 2 blocks of 12 bits each and 2 blocks of 6 bits each. It is spread over 22 bursts in the following way :

• the first and the twenty-second bursts carry one block of 6 bits each
• the second and the twenty-first bursts carry one block of 12 bits each
• the third and the twentieth bursts carry one block of 18 bits each
• from the fourth to the nineteenth burst, a block of 24 bits is placed in each burst

 
A burst will then carry information from five or six consecutive data blocks. The data blocks are said to be interleaved diagonally. A new data block starts every four bursts.

 

    Burst assembling

The busrt assembling procedure is in charge of grouping the bits into bursts. Section 5.2.3 presents the different bursts structures and describes in detail the structure of the normal burst.
 

 
   Ciphering
 
Ciphering is used to protect signaling and user data. First of all, a ciphering key is computed using the algorithm A8 stored on the SIM card, the subscriber key and a random number delivered by the network (this random number is the same as the one used for the authentication procedure). Secondly, a 114 bit sequence is produced using the ciphering key, an algorithm called A5 and the burst numbers. This bit sequence is then XORed with the two 57 bit blocks of data included in a normal burst.

In order to decipher correctly, the receiver has to use the same algorithm A5 for the deciphering procedure.

 

   Modulation

The modulation chosen for the GSM system is the Gaussian Minimum Shift Keying (GMSK).

The aim of this section is not to describe precisely the GMSK modulation as it is too long and it implies the presentation of too many mathematical concepts. Therefore, only brief aspects of the GMSK modulation are presented in this section.

The GMSK modulation has been chosen as a compromise between spectrum efficiency, complexity and low spurious radiations (that reduce the possibilities of adjacent channel interference). The GMSK modulation has a rate of 270 5/6 kbauds and a BT product equal to 0.3. Figure 5 presents the principle of a GMSK modulator.

 

  figure 5: GMSK modulator

 

 
Discontinuous transmission (DTX)

This is another aspect of GSM that could have been included as one of the requirements of the GSM speech codec. The function of the DTX is to suspend the radio transmission during the silence periods. This can become quite interesting if we take into consideration the fact that a person speaks less than 40 or 50 percent during a conversation. The DTX helps then to reduce interference between different cells and to increase the capacity of the system. It also extends the life of a mobile's battery. The DTX function is performed thanks to two main features:

• The Voice Activity Detection (VAD), which has to determine whether the sound represents speech or noise, even if the background noise is very important. If the voice signal is considered as noise, the transmitter is turned off producing then, an unpleasant effect called clipping.
• The comfort noise. An inconvenient of the DTX function is that when the signal is considered as noise, the transmitter is turned off and therefore, a total silence is heard at the receiver. This can be very annoying to the user at the reception because it seems that the connection is dead. In order to overcome this problem, the receiver creates a minimum of background noise called comfort noise. The comfort noise eliminates the impression that the connection is dead.

   

Timing advance

The timing of the bursts transmissions is very important. Mobiles are at different distances from the base stations. Their delay depends, consequently, on their distance. The aim of the timing advance is that the signals coming from the different mobile stations arrive to the base station at the right time. The base station measures the timing delay of the mobile stations. If the bursts corresponding to a mobile station arrive too late and overlap with other bursts, the base station tells, this mobile, to advance the transmission of its bursts.

 

Power control

At the same time the base stations perform the timing measurements, they also perform measurements on the power level of the different mobile stations. These power levels are adjusted so that the power is nearly the same for each burst.

A base station also controls its power level. The mobile station measures the strength and the quality of the signal between itself and the base station. If the mobile station does not receive correctly the signal, the base station changes its power level.

 

Discontinuous reception

It is a method used to conserve the mobile station's power. The paging channel is divided into subchannels corresponding to single mobile stations. Each mobile station will then only 'listen' to its subchannel and will stay in the sleep mode during the other subchannels of the paging channel.

 

Multipath and equalisation

At the GSM frequency bands, radio waves reflect from buildings, cars, hills, etc. So not only the 'right' signal (the output signal of the emitter) is received by an antenna, but also many reflected signals, which corrupt the information, with different phases.

An equaliser is in charge of extracting the 'right' signal from the received signal. It estimates the channel impulse response of the GSM system and then constructs an inverse filter. The receiver knows which training sequence it must wait for. The equaliser will then , comparing the received training sequence with the training sequence it was expecting, compute the coefficients of the channel impulse response. In order to extract the 'right' signal, the received signal is passed through the inverse filter.

 
 

GSM services

It is important to note that all the GSM services were not introduced since the appearance of GSM but they have been introduced in a regular way. The GSM Memorandum of Understanding (MoU) defined four classes for the introduction of the different GSM services:

• E1: introduced at the start of the service.
• E2: introduced at the end of 1991.
• Eh: introduced on availability of half-rate channels.
• A: these services are optional.

 
Three categories of services can be distinguished:

• Teleservices.
• Bearer services.
• Supplementary Services.

 

Teleservices
 

- Telephony (E1® Eh).

- Facsmile group 3 (E1).

- Emergency calls (E1® Eh).

- Teletex.

- Short Message Services (E1, E2, A). Using these services, a message of a maximum of 160 alphanumeric characters can be sent to or from a mobile station. If the mobile is powered off, the message is stored. With the SMS Cell Broadcast (SMS-CB), a message of a maximum of 93 characters can be broadcast to all mobiles in a certain geographical area.

- Fax mail. Thanks to this service, the subscriber can receive fax messages at any fax machine.

- Voice mail. This service corresponds to an answering machine.

 

Bearer services

A bearer service is used for transporting user data. Some of the bearer services are listed below:

• Asynchronous and synchronous data, 300-9600 bps (E1).
• Alternate speech and data, 300-9600 bps (E1).
• Asynchronous PAD (packet-switched, packet assembler/disassembler) access, 300-9600 bps (E1).
• Synchronous dedicated packet data access, 2400-9600 bps (E2).

 

Supplementary Services

- Call Forwarding (E1). The subscriber can forward incoming calls to another number if the called mobile is busy (CFB), unreachable (CFNRc) or if there is no reply (CFNRy). Call forwarding can also be applied unconditionally (CFU).

- Call Barring. There are different types of `call barring' services:

• Barring of All Outgoing Calls, BAOC (E1).
• Barring of Outgoing International Calls, BOIC (E1).
• Barring of Outgoing International Calls except those directed toward the Home PLMN Country, BOIC-exHC (E1).
• Barring of All Incoming Calls, BAIC (E1)
• Barring of incoming calls when roaming (A).

- Call hold (E2). Puts an active call on hold.

- Call Waiting, CW (E2). Informs the user, during a conversation, about another incoming call. The user can answer, reject or ignore this incoming call.

- Advice of Charge, AoC (E2). Provides the user with an online charge information.

- Multiparty service (E2). Possibility of establishing a multiparty conversation.

- Closed User Group, CUG (A). It corresponds to a group of users with limited possibilities of calling (only the people of the group and certain numbers).

- Calling Line Identification Presentation, CLIP (A). It supplies the called user with the ISDN of the calling user.

- Calling Line Identification Restriction, CLIR (A). It enables the calling user to restrict the presentation.

- Connected Line identification Presentation, CoLP (A). It supplies the calling user with the directory number he gets if his call is forwarded.

- Connected Line identification Restriction, CoLR (A). It enables the called user to restrict the presentation.

- Operator determined barring (A). Restriction of different services and call types by the operator.

 
 

Conclusion

The aim of this page was to give an overview of the GSM system and not to provide a complete and exhaustive guide.

As it is shown in this page, GSM is a very complex standard. It can be considered as the first serious attempt to fulfil the requirements for a universal personal communication system. GSM is then used as a basis for the development of the Universal Mobile Telecommunication System (UMTS).

 

<Prev>