Bad CSSR after the deactivation of "TMSI Reallocation"

[gogotchiya83]

Dear Expert;

In order to improve our Call Establishment Time, we deactivate the TMSI Reallocation in core side,
But unfortunately after this action we had a degradation in our CSSR (aproximately 0,2%).

Thanks to explain us why we had this impact, and if we have to make some tuning in order to improve the CSSR.

Regards//
Gogo

[babak1349]

Hi there,
I think you need to check your paging success rate. Maybe it has affected your paging success rate as well. when you are saying CSSR, from which node point of view it is ? is from Radio part ?or Core Network KPI? In Radio part normally it does consider Air Interface signalling (SDCCH or RRC) and Data part allocation (TCH assignment or RAB assignment) But in core it might consider the paging as well. normally in Radio KPIs the paging success rate is not considered in CSSR KPi calculation
Anyway 0.2% is too small and it is not easy to detect the cause

Cheers

Dear Expert;

In order to improve our Call Establishment Time, we deactivate the TMSI Reallocation in core side,
But unfortunately after this action we had a degradation in our CSSR (aproximately 0,2%).

Thanks to explain us why we had this impact, and if we have to make some tuning in order to improve the CSSR.

Regards//
Gogo

[gogotchiya83]

The Degradation is seen from Radio Part especially The indiactor "TCH_Setup_Success_Rate", for the Paging SR no changes after the deactivation of TMSI Reallocation ...

[babak1349]

Hi,
I think you are talking about GSM network ? Right ?
As far as I know TMSI is used for paging and maybe for Authentication I understood from what you said, TCH assignment has been affected (even though it is too small) it is better you check in cell level to see all cells are affected or just some specific cells are affected and see what is the reason of TCH assignment failure But I think TCH assignment failure noting to do with TMSI So maybe it is some users causing problem in some specific cells

Cheers



The Degradation is seen from Radio Part especially The indiactor "TCH_Setup_Success_Rate", for the Paging SR no changes after the deactivation of TMSI Reallocation ...

[electron]

Hi,

TMSI Reallocation does not have any impact on SCCP time it is basically just to improve confidentiality also 0.2% is negligible and even could be an impact of cell level issue.
Dear Expert;

In order to improve our Call Establishment Time, we deactivate the TMSI Reallocation in core side,
But unfortunately after this action we had a degradation in our CSSR (aproximately 0,2%).

Thanks to explain us why we had this impact, and if we have to make some tuning in order to improve the CSSR.

Regards//
Gogo

[gogotchiya83]

Thanks for your reply,

the impact is seen on the Whole Network and the number of Failed Assignments due to (Clear Commands Sent By MSC) increased after the deactivation of the TMSI Reallocation...

Any explanation of this issue?

[babak1349]

Hi,
It is weird. what could be relation between TMSI reallocation and TCH assignment failure ? TMSI is used for users' Authentication and recognition I am not sure maybe it is used for ciphering as well. Do you have ciphering activated in your network ? Why you don't suggest to re-activate TMSI reallcation for worst MSC and see the impact?
Can you check as well SDCCH drop to see if it has increased ?
What I noticed the TMSI relaaocation is done during Location update (if it s not successful user can not make any call at all) and also TMSI rellocation is a part of call setup but if it fails SDCCH drop should increase So here is further suggestion:
1) Check your SDCCH accessibility did it change ?
2) check SDCCH drop . did it change? it is important to check it because if any signaling procedure during call setup fails you will see it in SDCCH drop
3) check TCH assignment success rate
4) check your CSSR formula

Correct CSSR formula should be like this : (SDCCH success rate For CS CALLs)X (1-SDCCH drop)X( TCH assignment success rate) so you need to check which part of the formula has been degraded ( depending on your defined formulas)
SC I am not sure your interperting correctly . Please Note Assignment Request is coming from core and Assignment Response comes from BSC So actually the TCH assignment procedure will not start. If there is TCH assignment Request from core then Failure message should come from BSS So what is the case ? after TCH assignment Response Core sends clear message ?

Cheers

[gogotchiya83]

Dear babak1349

Thakns for your reply,

AS i explain before, no change was seen in SD Drop/Sd Cong and the impact is on the whole Network

At the same date when we dactivate the TMSI Reallocation we had an increase o the number of Failed Assignments (Clear Commands Sent By MSC) (A3129H) which cause a degradation of the TCH ASS SR (The same CSSR formula as u suggest)...

Regards//
Gogo

[RF engineer]

Dear babak1349

Thakns for your reply,

AS i explain before, no change was seen in SD Drop/Sd Cong and the impact is on the whole Network

At the same date when we dactivate the TMSI Reallocation we had an increase o the number of Failed Assignments (Clear Commands Sent By MSC) (A3129H) which cause a degradation of the TCH ASS SR (The same CSSR formula as u suggest)...

Regards//
Gogo

CHECK BELOW LINK IT MAY HELP YOU
http://www.finetopix.com/showthread.php?33152-Huawei-ATCA-MSC-majic-by-improving-CSSR-on-cell&highlight=ATCA

[gogotchiya83]

We didn't change any timer, the only action done is the deactivation of the TMSI Reallocation which increase the number of Failed ASS du to CLR CMD From MSC...

Any Help PLZ

[RF engineer]

So rollback change done on MSC in order to see if it back to normal situation or Not.

[ahmed planning]

dear gogtchiya83

we are using huawei BSC 6900 .
please can you show me the IMM ASS formula with counter IDs of your vendors

[gogotchiya83]

We have HW Equipement (BSS & NSS) (GBSS 13.0/BSC 6900)
as it was noticed before, we had an increase of the number of Failed Assignments due to (Clear Commands Sent By MSC) (counter : A3129H)

any help?

[babak1349]

Hello Gogotechiya83,
I searched on the net to find out more about that clearing message Here is the link of a doc on scribd about that type of failure

http://www.scribd.com/doc/127009973/Investigation-Report-for-TCH-Assignment-Failure-20120822

What is all it saying in some cases the UE sends disconnect message before the Assignment process gets completed It is pure again NAS massages
Other case it may happen is when the MSC doesn't receive Assignment Complete message and it sends clearing message to BSc to release the resources In this case it could be bad quality in UE side or TN issue or MSC sends the clear message too quick (why it may happen maybe the timer related to Assignment process in MSC expires too quickly)

Anyway all those causes is nothing to do with TMSI reallocation That failure happens during TCH assignment procedure and also with UE
do you have ciphering activated in your network ?
Can you check with traces to see if you have "disconnect" message received from UE and followed by "release" command from MSC... Maybe you can see disconnect reason ....

Cheers


We have HW Equipement (BSS & NSS) (GBSS 13.0/BSC 6900)
as it was noticed before, we had an increase of the number of Failed Assignments due to (Clear Commands Sent By MSC) (counter : A3129H)

any help?

[gogotchiya83]

Thanks Babak for your Hepl,

We have the "Ciphering" activated, we will try to make some traces with core engineers in order to understand the reason of this failures...

[babak1349]

hi,
you welcome. Thanks for highlighting the issues and this way we can learn more
As that document explained the problem could be UEs or it could be in core network site when it sends the "release command" early So activating trace could be a good idea. It doesn't look like that BSC sends any failure as in this case BSC should send RAS assignment response with failure cause (for example if there is lack of resources)
Please let us know the progress Maybe it would be good idea to run traces over A interface because you can collect more data.

Just another point, I know there no changes in parameters But there isa timer in MSC side which is T10 and used to supervise TCH Assignment process and if it fails, the MSC sends clear message So you sohuld check that parameter as well to see in any case if change happened

Cheers

Thanks Babak for your Hepl,

We have the "Ciphering" activated, we will try to make some traces with core engineers in order to understand the reason of this failures...