8.25 Overload Control

25.4133GPPRelease 17TSUTRAN Iu interface Radio Access Network Application Part (RANAP) signalling

8.25.1 General

This procedure is defined to give some degree of signalling flow control. At the UTRAN side, "Processor Overload" and "Overload in the Capability to Send Signalling Messages to the UE" are catered for, and at the CN side, "Processor Overload" is catered for. The procedure uses connectionless signalling.

The philosophy used is to stem the traffic at source with known effect on the service. The algorithm used is:

At the CN side:

– If TigOC is not running and an OVERLOAD message or "Signalling Point Congested" information is received, the traffic should be reduced by one step. It is also possible, optionally, to indicate the number of steps to reduce the traffic within the Number of Steps IE. At the same time, timers TigOC and TinTC should be started.

– During TigOC all received OVERLOAD messages or "Signalling Point Congested" information should be ignored.

– This step by step reduction of traffic should be continued until maximum reduction is obtained by arriving at the last step.

– If TinTC expires, the traffic should be increased by one step and TinTC should be re-started unless the number of steps by which the traffic is reduced is back to zero.

At the UTRAN side:

– If TigOR is not running and an OVERLOAD message not including the Priority Class Indicator IE or "Signalling Point Congested" information is received, all the traffic should be reduced by one step. It is also possible, optionally, to indicate the number of steps to reduce the traffic within the Number of Steps IE. At the same time, timers TigOR and TinTR should be started.

– If TigOR is not running and an OVERLOAD message including the Priority Class Indicator IE is received and a procedure for reduction of all traffic is not being processed, then signalling traffic for the indicated priority class should be reduced by one step. It is also possible, optionally, to indicate the number of steps to reduce the traffic within the Number of Steps IE. At the same time, timers TigOR and TinTR should be started.

– During TigOR all received OVERLOAD messages or "Signalling Point Congested" information should be ignored.

– This step-by-step reduction of traffic should be continued until maximum reduction is obtained by arriving at the last step.

– If TinTR expires, the traffic should be increased by one step and TinTR should be re-started unless the number of steps by which the traffic is reduced is back to zero.

The number of steps and the method for reducing the load are implementation-specific.

There may be other traffic control mechanisms from O&M activities occurring simultaneously.

8.25.2 Philosophy

Void.

8.25.3 Successful Operation

8.25.3.1 Overload at the CN

Figure 27: Overload at the CN. Successful operation.

The CN should indicate to the RNC that it is in a congested state by sending an OVERLOAD message. The CN Domain Indicator IE may be included, if the CN can determine the domain suffering from the signalling traffic overload. A specific CN node shall send this message only towards those RNCs from which it can receive the INITIAL UE MESSAGE message.

Reception of the message by the UTRAN should cause reduction of signalling traffic towards the CN. If the CN Domain Indicator IE is included in the OVERLOAD message, and the Global CN-ID IE is not, the RNC should apply signalling traffic reduction mechanisms towards the indicated domain. If the Priority Class Indicator IE is included then the RNC should take the appropriate action to reduce the traffic for the priority class indicated in the IE.

If the NNSF is active, the CN shall include the Global CN-ID IE within the OVERLOAD message, and the RNC should apply signalling traffic reduction mechanisms towards the indicated CN node only.

8.25.3.2 Overload at the UTRAN

Figure 28: Overload at the UTRAN. Successful operation.

If the UTRAN is not capable of sending signalling messages to UEs due to overloaded resources, the UTRAN should send an OVERLOAD message to the CN. The RNC shall include the Global RNC-ID IE in this message. The message shall be sent only towards those CN nodes towards which the RNC can send the INITIAL UE MESSAGE message. If the Priority Class Indicator IE is included it shall be ignored.

8.25.4 Abnormal Conditions

Not applicable.