FAQ on RRC (Radio Resource Control Layer)

By Zahid Ghadialy ([email protected])

Last Updated: 01/10/2005

For SIBs to be transmitted on physical channels, segmentation ofSIB blocks may or may not be needed (because of the fixed size), Sowhere is this segmentation actually done ?? .. at Node B or RNC?

The segmentation is done in RNC (CRNC). 25.430 says

Section 5.2.4 System Information Management

System Information is sent by the CRNC to a Node B. CRNC can alsorequest the Node B to autonomously create and update certain Node Brelated system information. Scheduling of system broadcastinformation is carried out in the CRNC. Scheduling information isalways sent by the CRNC to the Node B. The Node B is responsible fortransmitting the received system information according to thescheduling parameters provided. If requested by the CRNC, the Node Bis also responsible for autonomously creating and updating the Node Brelated system information according to the scheduling parametersprovided.

There are 18 types of SIBs in all, Is there any SIB that isindependently generated at Node B, and is sent to UE, without anyRNCs involvement.?

Sib 7 is generated by Node B independent of RNC

Between Node B and RNC, NBAP Protocol is used, is all theInformation Elements (IE) transmitted from RNC to Node B areASN.1 Encoded? At Node B, is Decoding of these IEs is not requiredand can be sent directly to UE?

The elements sent from RNC to Node B are already encoded. They areencoded by the RRC layer in RNC. They are sent directly by Node Bwithout need for decoding. Since Sib 7 is generated by Node B, it has tocall a method in RNC to encode Sib 7 and then it is transmitted by NodeB.

Annex D in 25.433 shows how data is first split into segments and thenRRC encoding is performed on those segments. These F and V (Fixedand Variable size) segments are sent to Node B over Iub inIB_SG_DATA IE in SYSTEM INFORMATION UPDATE REQUESTmessage.

When does the UE send "Physical Channel ReconfigurationFailure" message with the cause set to "compressed mode runtimeerror"? What is the UTRAN expected to do afterwards?

FAQ on RRC (Radio Resource Control Layer) http://www.3g4g.co.uk/Faq/faq0003_rrc.html

1 of 5 1/27/2011 3:45 PM

The UE sends "Physical Channel Reconfiguration Failure" message withthe cause set to "compressed mode runtime error" in case when morethan one compressed mode patterns are setup. If the compressed modepatterns create an illegal overlap than this error should be reported to theUTRAN.

The following is extract from 25.331***************8.2.11.2 Runtime error due to overlapping compressed modeconfigurationsWhen the UE has received from the UTRAN the configurations ofseveral compressed mode transmission gap pattern sequences, and ifseveral of these patterns are to be simultaneously active, the UE shallcheck to see if these simultaneously active transmission gap patternsequences create transmission gaps in the same frame. An illegal overlapis created if two or more transmission gap pattern sequences createtransmission gaps in the same frame, irrespective of the gaps are createdin uplink or downlink.If the parallel transmission gap pattern sequences create an illegaloverlap, the UE shall:1> delete the overlapping transmission gap pattern sequenceconfiguration stored in the variable TGPS_IDENTITY, which isassociated with the highest value of IE "TGPSI";1> transmit a PHYSICAL CHANNEL RECONFIGURATIONFAILURE message on the DCCH using AM RLC, setting theinformation elements as specified below:2> not include the IE "RRC transaction identifier";2> set the cause value in IE "failure cause" to value "compressed moderuntime error".1> terminate the inter-frequency and/or inter-RAT measurementscorresponding to the deleted transmission gap pattern sequence;1> when the PHYSICAL CHANNEL RECONFIGURATION FAILUREmessage has been submitted to lower layers for transmission:2> the procedure ends.*****************************

UTRAN has O&M logging facility for logging everything. It will also logall these kind of errors and the engineers that analyse the logs will fixthis problem in the network. In a stable network this problem cannotoccur.

How does the initial UE Cell Selection takes place?

The initial cell-selection procedure is used in case there is no informationon the current environment stored in the UE. However, normally the UEstarts the cell selection with a stored information cell-selectionprocedure. The UE may have stored the necessary information of thecell it was previously camped on, such as frequency and scramblingcode. The UE may first try to synchronize into that cell, and if it fails, itmay trigger the initial cell selection.

The purpose of the initial cell-selection procedure is to find a cell, notnecessarily the best cell, but a usable cell, for the UE to camp on afterpower-on. In the UTRAN, the number of carrier frequencies is quitesmall. One operator typically operates only on two or three frequency


2 of 5 1/27/2011 3:45 PM

carriers. In the first phase of UMTS in Europe, the frequency allocationfor UMTS-FDD is 2 60 MHz (uplink/downlink), which means thatthere can be, at most, only 12 carrier frequencies of 5-MHz bandwidtheach. These carriers are then divided between up to six operators. Eachcarrier will only support one operator. This obviously forces theoperators to coordinate their networkplanning activities near nationalborders because the same frequency can be used by different operatorsin adjacent countries.

The specifications do not accurately dictate how the initial cellselectionprocedure should be implemented; it is left for the UE manufacturers todecide. Most of the functionality, however, has to be in the physicallayer, and the RRC layer has only a management role. The initialcell-selection procedure is performed on one carrier frequency at a timeuntil a suitable cell is found. In principle the process includes thefollowing:

Search for primary synchronization channels (P-SCHs);1.Once such a channel is found, acquire time-slot synchronizationfrom it;

2.

Acquire frame synchronization from the corresponding S-SCH;3.Acquire the primary scrambling code from the correspondingCPICH;

4.

Decode system information from the cell to check whether it is asuitable cell for camping (i.e., it contains the right PLMN code andaccess to it is allowed).

5.

All P-SCHs have the same fixed primary synchronization code. Thesearch procedure should yield a set of P-SCHs in the area. Because theP-SCH is only transmitted during the first 256 chips of each time slot,the beginning of its transmission also indicates the start of a time slot inthe corresponding cell.

In the second phase of the process, the received signal is correlated withall possible secondary synchronization code (S-SCH) words on theS-SCH. There are 16 different SSCs, and these can be combined into 64different code words, each with a length of 15 SSCs. Once the right codeword is found, this gives the UE the frame synchronization and the codegroup identity, which indicates eight possible primary scrambling codesfor the control channels.

The third phase of the procedure consists of finding the right primaryscrambling code for this cell. Each candidate cells primary scramblingcode (there are eight of them as shown in the second phase) is applied, inturn, to the common pilot channel (CPICH) of that cell. Because theCPICH carries a predefined bit/symbol sequence, the UE knows when ithas found the correct primary scrambling code. The resolved primaryscrambling code can then be used to detect the CCPCH, which carriesthe BCH, which contains the system information the UE is seeking.There are various ways to optimize this procedure to make it quicker.Note that phase five actually contains another major procedure, PLMN(i.e., the operator) selection. PLMN is identified by a PLMN code, anumber that is transmitted on the BCCH channel of that network. A UEtries to find its home PLMN, the operator it has a contract with. Inprinciple, a UE should first scan through all UTRAN frequencies until a


3 of 5 1/27/2011 3:45 PM

good PLMN is found, and then start an initial cell-selection process onthat frequency.

Note that one frequency can only be used by one operator (except inareas near country borders). However, while looking for the right PLMNcode, the UE has already obtained all the necessary information forcamping on a suitable cell, and no new scanning procedure is necessaryonce the correct PLMN is found. The situation is different if the UE isroaming abroad, and the home PLMN is not found. In that case RRC hasto report all available PLMNs to NAS and wait for its selection decision,which can be either automatic or manual (user selection). This is timeconsuming, and many readers may have noticed this phenomenon whenarriving at an airport in a new country and switching their GSM phoneson. It may take a very long time before the phone registers to a network,especially if the phone is a multimode model with several frequencybands to scan.

The initial cell-selection process is repeated as many times as necessaryuntil the first suitable cell is found for camping. Once the UE hasmanaged to camp on a cell, it decodes the system information from it,including the neighbor cell list. This information can be used to help theUE find the best cell to camp onto. Note that the initial cell-selectionprocedure only found a cell to camp on (the first possible cell). It ispossible that this cell will not be the best possible cell. For example,there could have been other frequencies including better cells for thisparticular UE that had not yet been scanned.

The neighbor cell list immediately tells the UE which frequencies andneighbor cells should be checked while the best possible cell is beingsearched for. The list includes additional information that can be used tooptimize the cell-synchronization procedure, information such as theprimary scrambling codes and timing information (optional, relative tothe serving cell). With this information it should be possible to quicklydescramble the CPICH from a neighbor cell.

From the CPICH it is possible to calculate the received chip energy to-noise ratio (Rx Ec/No) for this cell. This measurement is acquired foreach neighbor cell in the list. Based on this information, the UE candetermine whether there are better cells available. From a possiblecandidate cell, the UE must decode the system information to check thatit is not barred for access.

If the neighbor cell list contains cells from another RATfor example,GSM cellsand the serving cell quality level is worse than the Ssearchparameter, then the GSM cells must be taken into consideration in thecell reselection procedure.

Reference:Introduction to 3G Mobile Communications - Juha Korhonen3GPP TS 25.304: UE Procedures in Idle Mode and Procedures for CellReselection in Connected Mode.3GPP TS 23.122: NAS Functions Related to Mobile Station (MS) In IdleMode.


4 of 5 1/27/2011 3:45 PM

Documents

FAQ on RRC (Radio Resource Control Layer)