Double BA Lists in GSM

The Double BA Lists feature is defined per cell. This means that it is possible to have double lists in one cell and a single list in another cell. A single list corresponds to having the same list in idle mode and in active mode.The Absolute Radio Frequency Channel Numbers (ARFCNs) to be measured on by an MS in a cell shall be included in the BA list and they are defined by the parameter MBCCHNO. Only ARFCNs for the neighboring cells BCCH frequencies shall be defined (serving cells BCCH frequency can also be included if separate idle mode and active mode BA lists are used). With the parameter LISTTYPE it is possible to determine whether the list is for idle mode or active mode, or both.
Basically up to 32 measurement frequencies can be defined in the BA list. If UTRAN neighbors are registered in the cell, then the maximum number of frequencies in the BA list is 31.The following restrictions will apply according to the global system type and the list type. Note that if the list type is not specified, it is set to BOTH. When the list type is set to BOTH, the restrictions are checked for both IDLE and ACTIVE lists.

Transfer of BA Lists to the Mobile Stations

An MS in idle mode receives the idle mode BA list in System Information 2, 2bis and 2ter messages that are transmitted on BCCH. The format and scheduling is described in the 3GPP Technical Specification 44.018.

An MS in active mode receives the active mode list in System Information 5, 5bis and 5ter messages transmitted on the SACCH.

Measurement Reports

In active mode the MS reports the signal strength and estimated quality of the serving cell and the signal strength of the six strongest neighbours, for which the BSIC is decoded. The measurement reports are sent on the SACCH every SACCH period,  The Measurement Reports contain the downlink measurements, information about the measured frequencies and BSIC for the reported cells and information about what list (old or new) that has been used,

If EMR is activated then the MS may send Enhanced Measurement Reports instead. An Enhanced Measurement Report is a new measurement report containing additional data, of which Bit Error Probability (BEP) and the number of correctly received speech frames are the two most important ones.

The Active Mode BA List

The active mode BA list should correspond to the defined neighbors. The BCCH carrier of serving cell should not be in the active mode list. A short active mode BA list gives better handover performance than a long list. If a long active mode list is used there will be less samples taken from each cell, thus resulting in decreased measurement accuracy. It could also be the case that the MSs spends time decoding the BSIC of an irrelevant cell. Furthermore, if this is not successful it will spend extra time trying to decode the BSIC of this cell and waits longer time before it re-decodes the BSICs of the other cells. Thus, this increases the risk that they are no longer valid. The active mode list should, in the normal case, not exceed 15 frequencies.

The Idle Mode BA List

The idle mode BA list should normally correspond to all the BCCH carriers used. Serving cell should be included in the idle mode BA list. When an MS is switched on, the MS scans the latest received BA list. If the MS has moved, it could be the case that the MS measures a very weak signal from one of the cells in the old list, and camps on it. If this list is very short and does not contain the cells which are the best ones in the current location of the MS, the MS will continue to camp on the "wrong" cell. To set up a call under these circumstances is not favorable. This situation can be avoided if the idle mode list contains more frequencies. Many frequencies in the idle mode list means that less samples are taken per frequency than what would have been the case if the list was short. The decreased accuracy for the idle mode measurements that this gives rise to is not considered a problem.

Main Controlling Parameters

MBCCHNO is the list of BCCH carriers, given as Absolute Radio Frequency Channel Numbers (ARFCNs), of the neighboring cells (and possibly serving cell) on which the MS shall measure. The parameter is set per cell and type of list (idle or active).

LISTTYPE is used to specify to which list, IDLE or ACTIVE, the stated frequencies belong when changes in the BA lists are made. If LISTTYPE is omitted the BCCH carriers defined or changed by MBCCHNO are valid in both lists.

Dynamic Half Rate Allocation (DHA)

The DHA feature optimizes the usage of capacity when the cell load is high, whilst offering the best possible speech quality when the cell load is low. This is achieved by allocating FR or HR TCHs in accordance with the cell load, at the time when a new TCH shall be selected due to assignment and most types of handovers. At high cell load HR TCHs have precedence and at low cell load FR TCHs have precedence.

The feature is invoked if the parameter DHA is ON. 

The parameter is set on cell level. The feature is not invoked at immediate assignment if using a TCH. Then it is the setting of the parameter CHAP that controls whether FR and/or HR may be used

To trigger the functionality it must also be checked that the DL signal strength measured by the MS is not too low. If the signal strength is below DHASSTHRASS (if during assignment) or DHASSTHRHO (if during handover) then dynamic HR allocation evaluation will not be performed. This signal strength check can be activated and deactivated using the parameter DHASS.

The thresholds for when allocation of HR TCHs is triggered are given by the parameters DTHAMR and DTHNAMR and are set per cell. These parameters are percentage values and are compared to the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs. The two parameters indicate that AMR and non AMR DR capable MSs may, depending on the cell load, be treated differently in the allocation of TCHs. If using the feature Speech Quality Priority, and turning the parameter DHPR ON, it is also possible to set these two parameters differently for different priority levels. By doing that it is for example possible to start allocating HR to low priority users at moderate cell load, and let high priority users get FR until the load in the cell is very high.

In general the following cases occur:

§  If the MS and the cell support AMR/HR and the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs is equal to or above the value of DTHAMR set for the specific priority level that the MS is assigned, then FR TCHs will have precedence over HR TCHs at channel allocation. If the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs is less than DTHAMR , but higher than DTHNAMR, then AMR/HR TCHs will have precedence over FR TCHs. If the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs is less than both DTHAMR and DTHNAMR, then HR TCHs will have precedence over FR TCHs (both AMR/HR and HR will have precedence, and in this order).

§  If the MS or the cell do not support AMR/HR and the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs is equal or above DTHNAMR set for the specific priority level that the MS is assigned, then FR TCHs will have precedence over HR TCHs. If the number of idle TCH BPCs divided by the total number of de-blocked TCH BPCs is less than DTHNAMR, then HR TCHs will have precedence over FR TCHs (only HR SPV1).

Log-Normal Fade Margins

The Log-normal fade margin is incorporated in to the link budget calculation to ensure coverage reliability, and there are three environmental cases to consider:

· Outdoor fade margin

· Indoor fade margin

· In-vehicle fade margin

Outdoor Fade Margin

The outdoor standard deviation α0, depends on the terrain: dense urban, urban, or suburban and may change very slightly with frequency. The outdoor standard deviation, α0, ranges from 5 dB (rural) to 12 dB (dense urban) with a typical value of 8 dB (suburban) Radio Wave Propagation A

Indoor Fade Margin

If an MS is inside a building, the received signal is attenuated as it passes through the exterior of the building. Building penetration loss, Lbp is subject to random variation. We denote the standard deviation of the building penetration loss as αbp. The standard deviation, α, used to compute the indoor fade margin accounts for the outdoor environment, as well as the random variation of the building penetration loss. The standard deviation, α 0, is pooled with α bp. Thus the pooled deviation,

In-vehicle Fade Margin

The in-vehicle loss is also variable; therefore, the standard deviation, α , should account the outdoor environment, as well as the random variation of the in-vehicle penetration loss. We denote the standard deviation of the in-vehicle penetration loss as αiv. The standard deviation, _iv, must be  pooled with the outdoor standard deviation αiv. The pooled standard

Cell level Traffic Load Counters for EDGE Evolution

TBFDCDLCAP : Number of downlink TBFs where the MS is capable of using dual carriers.

TRAFDCDLTBF :  Number of downlink TBFs, in EGPRS mode, reserved on dual carriers.

MAXDCTSDL : Maximum possible number of time slots reservable for MSs on downlink TBFs in EGPRS mode, reserved on dual carriers.

MUTILDCDL : Sum of percentage shares of reserved time slots for all EGPRS mode downlink TBFs reserved on dual carriers related to the maximum possible reservable time slots.

TRAFEEVOSCAN : Number of scans for the counters in this object type. This counter is only valid for counters in object type TRAFEEVO.

TSDCDL: Number of time slots with one or more uplink or downlink TBFs currently reserved on dual carriers.

Tower Mounted Amplifiers TMA

A Tower Mounted Amplifier (TMA) has a low noise figure and some gain. If it is mounted as closely as possible to the receive antenna, it can improve uplink sensitivity at the antenna, thus increasing the cell radius. There will be an improvement if the radio link is truly noise limited. If the system is uplink interference limited, there will be no improvement using a TMA. A TMA can degrade performance if the gain is set too high. There will be intermodulation effects due to larger signals, such as terminals close to the base station. Best results are achieved when the TMA gain is used to compensate for the feeder loss.

Frequency Reuse Patterns in GSM

Proper frequency planning is essential in the development of a quality cellular system. Due to the limited amount of available RF spectrum, the scarcity of channels available to each operator, and the fact that most available frequencies are non-contiguous for most operators, frequencies must be reused throughout the system to increase network capacity. The frequencies are assigned such that there is minimal cochannel and adjacent channel interference between sites. Frequency reuse is based on hexagonal cell groupings called clusters. The size of the cluster will determine how the cluster is repeated throughout the network, i.e. the reuse pattern. The frequency reuse patterns are designated as N/F. Where N is the number of cell sites in a cluster and F is the number of frequency groups within a cluster. Ericsson uses 7/21 and 4/12 reuse patterns. The cluster patterns

 In a 7/21 plan, there are 7 cell sites (A, B, . . . , G) and 21 frequency groups (A1, A2, A3, . . . , G1, G2, G3). Because A1 and G3 are adjacent frequency groups, G2 and G3 are switched in the pattern so as to prevent interference. In a 4/12 pattern, there are 4 cell sites and 12 frequency groups. The frequency groups for these patterns are described in this document. There is no frequency reuse within a cluster. The voice channel group is used to assign a frequency group to a cell.

  

BER ( Bit Error Rate )

The environmental effects on the received signal produce interference and impairments in the form of Bit Error Rate (BER) . BER is defined as the ratio of the number of incorrect bits received versus the total number of bits. BER is estimated by the RBS on the reverse-link and by the MS on the forward-link. IS-136 does not state explicitly how the BER should be estimated but does specify the recommended accuracy for BER estimations. BER measures the effects of radio environment-introduced impairments which are discussed in the Speech Quality and Link Budget documents in this RF module.

                          Coding of Estimated  BER

Class BER	BER Interval (%)
0	BER% < 0.01
1	0.01 ≤ BER% < 0.1
2	0.1 ≤  BER% < 0.5
3	0.5 ≤  BER% < 1.0
4	1.0 ≤  BER% < 2.0
5	2.0 ≤  BER% < 4.0
6	4.0 ≤  BER% < 8.0
7	8.0 ≤  BER%