Failures after admission

[tsiz]

Hi All,

I wanted to know if you all can share some experiences to lower or fix failures after admission.

Sites have good T1 capacity and aal2 and VC's are dimensioned correctly. No aal2 congestion is observed but still very high failures after admission.

This is with E/// equipment.

Let me know if you have any suggestions!

[mulham_kh]

Hi All,

I wanted to know if you all can share some experiences to lower or fix failures after admission.

Sites have good T1 capacity and aal2 and VC's are dimensioned correctly. No aal2 congestion is observed but still very high failures after admission.

This is with E/// equipment.

Let me know if you have any suggestions!


Hi
I got the same problem but it was due to lack in Transmission resource
Regards

[tsiz]

Hi
I got the same problem but it was due to lack in Transmission resource
Regards

Yeah Mulham...mine has good resources but still I see a lot of failures after admissions.

[mulham_kh]

Yeah Mulham...mine has good resources but still I see a lot of failures after admissions.

If this problem is common for all sites maybe you have a problem with the bandwidth which it is dedicated between the RNC and SGSN
Regards

[salamanda1969]

Hi,

You may want to describe what kind of failures you are seeing. For e.g. call setup failures (call ended without announcement), call going straight to voicemail, congestion (from performance stats) or RAB setup failures.

[tsiz]

Hi,

You may want to describe what kind of failures you are seeing. For e.g. call setup failures (call ended without announcement), call going straight to voicemail, congestion (from performance stats) or RAB setup failures.

The failures I have is pegged with the incrementation of pmnoFailedafterAdm. The voice access fails are very high. retainability is pretty decent.

[zhanglw268]

Hi Tsiz,

pmnoFailedafterAdm can be pegged for SRB and RAB establishment, you can check certain stats to see whether it is related to SRB or RAB only, or they are mixed.
(1) RAB setup success rate check:
pmNoRabEstablishSuccess/pmNoRabEstablishAttempt, if they are nearly 100%, you can exclude RAB issue, then focus on the SRB stage
(2) SRB setup success rate check
If the pmnoFailedafterAdm happened at the SRB stage, the RRC success rate will be low, you can compare pmnoFailedafterAdm with
[pmTotNoRrcConnectReq - pmTotNoRrcConnectReqSuc - pmTotUtranRejRrcConnReq], if they're almost equal, then it is SRB.
(3). If both stages are involved, then (1) and (2) will not be true

The main check points shall be AAL2 setup and the procedure for NBAP "Radio Link setup" and "Radio link Reconfig".

Hope the above helps.

Regards

David

[edy.djunaedi]

Hi All,

I wanted to know if you all can share some experiences to lower or fix failures after admission.

Sites have good T1 capacity and aal2 and VC's are dimensioned correctly. No aal2 congestion is observed but still very high failures after admission.

This is with E/// equipment.

Let me know if you have any suggestions!


HI,

admission control is related to resource reservation..
resource on 3G is :
1. CE
2. Transport.
3. Power

so if you you didn't find the problem on transport, also check the CE utilization, check the power allocation..
and capability node B or RNC to guarantee the service like ; min or max bitrate related to parameter setting and required service bit rate.

you can find it by counter measuremet..

on procedure after pass the admission control by rnc it should be check the availability transport and then Node B (CE)

if there congest due to CE then propose to Upgrade CE or decrease max bitrate UL PS
if there overload due to power then adjust Ptx target and ptx Offset, Prx target and prx Offset.

CMIIW

BR//

[tsiz]

Hi Tsiz,

pmnoFailedafterAdm can be pegged for SRB and RAB establishment, you can check certain stats to see whether it is related to SRB or RAB only, or they are mixed.
(1) RAB setup success rate check:
pmNoRabEstablishSuccess/pmNoRabEstablishAttempt, if they are nearly 100%, you can exclude RAB issue, then focus on the SRB stage
(2) SRB setup success rate check
If the pmnoFailedafterAdm happened at the SRB stage, the RRC success rate will be low, you can compare pmnoFailedafterAdm with
[pmTotNoRrcConnectReq - pmTotNoRrcConnectReqSuc - pmTotUtranRejRrcConnReq], if they're almost equal, then it is SRB.
(3). If both stages are involved, then (1) and (2) will not be true

The main check points shall be AAL2 setup and the procedure for NBAP "Radio Link setup" and "Radio link Reconfig".

Hope the above helps.

Regards

David

Thanks David. For one ROP, I had the following values
pmTotNoRrcConnectReq-1250
pmTotNoRrcConnectReqSuc-615
pmTotUtranRejRrcConnReq-0
pmnoFailedafterAdm-28

So there is a big difference here. My CSSR and the CS RRC% are very low and the RAB% is very good.

[tsiz]

HI,

admission control is related to resource reservation..
resource on 3G is :
1. CE
2. Transport.
3. Power

so if you you didn't find the problem on transport, also check the CE utilization, check the power allocation..
and capability node B or RNC to guarantee the service like ; min or max bitrate related to parameter setting and required service bit rate.

you can find it by counter measuremet..

on procedure after pass the admission control by rnc it should be check the availability transport and then Node B (CE)

if there congest due to CE then propose to Upgrade CE or decrease max bitrate UL PS
if there overload due to power then adjust Ptx target and ptx Offset, Prx target and prx Offset.

CMIIW

BR//

Thanks Edy,

How do I calculate the UL CE utilization for the site? I tried pmapomcofullinkcap_avg but its not giving me in rnc or node b level.

The UE CE usage is 4829 with 3 sector and 3 carriers sharing 7 RaxB. Also where do I find the parameters Ptx target and ptx Offset, Prx target and prx Offset?

Thanks.

[zhanglw268]

Hi Tsiz,

From the statistics you provided, the pmnoFailedafterAdm is only about 2% of the total request, The RRC connection success Rate is around 50%, beside checking the transmission and CEs, Could you please check the below?
a). Is the cell at the 2 RNCs' border? If the cell is at the RNC border, the Mobile/UE may reselect to another cell in another RNC after sending the RRC setup request.
b). Use GPEH to check the Ec/No value when the UE sends RRC Connection Request, it looks like that you have a pilot polution or UL interference, or too big coverage for the cell.

Regards

David



Thanks David. For one ROP, I had the following values
pmTotNoRrcConnectReq-1250
pmTotNoRrcConnectReqSuc-615
pmTotUtranRejRrcConnReq-0
pmnoFailedafterAdm-28

So there is a big difference here. My CSSR and the CS RRC% are very low and the RAB% is very good.

[tsiz]

Hi Tsiz,

From the statistics you provided, the pmnoFailedafterAdm is only about 2% of the total request, The RRC connection success Rate is around 50%, beside checking the transmission and CEs, Could you please check the below?
a). Is the cell at the 2 RNCs' border? If the cell is at the RNC border, the Mobile/UE may reselect to another cell in another RNC after sending the RRC setup request.
b). Use GPEH to check the Ec/No value when the UE sends RRC Connection Request, it looks like that you have a pilot polution or UL interference, or too big coverage for the cell.

Regards

David

Thanks David. Yes the cells is in the RNC border and there is pilot pollution. Tilts have been changed but still there is accessibility issues. I know there would be a lot of signalling that would also cause the accessibility failures.

[edy.djunaedi]

Thanks David. For one ROP, I had the following values
pmTotNoRrcConnectReq-1250
pmTotNoRrcConnectReqSuc-615
pmTotUtranRejRrcConnReq-0
pmnoFailedafterAdm-28

So there is a big difference here. My CSSR and the CS RRC% are very low and the RAB% is very good.

Hi,

I think your problem is on RRC con Reject close to 50 %
RAB is not problem after other RRC conn attempt was success on 50 %

so my suggestion is :

1. change Rx Level Minimum and Rx Qual Minimum to support good radio condition when call start
ex : -115 become -110 (Rx lev min to camp on cell)
- 18 become - 15 (Rx Qual min to camp on cell)

2. increase transmit power of RACH but carefully cause can make UL interference.
3. check / change transmit power of CPICH to low level is use to min coverage but max performance ex 33 dBm to 30 dBm
4. check overload control due to power set transmit and receive power of cell (Ptx and Prx)

this is just clue to more focus on understanding your problem..
read your document related to 4 item above for their setting and their function related to RRC Conn Reject because Admision control not admited

CMIIW
BR//edy

[tsiz]

Hi,

I think your problem is on RRC con Reject close to 50 %
RAB is not problem after other RRC conn attempt was success on 50 %

so my suggestion is :

1. change Rx Level Minimum and Rx Qual Minimum to support good radio condition when call start
ex : -115 become -110 (Rx lev min to camp on cell)
- 18 become - 15 (Rx Qual min to camp on cell)

2. increase transmit power of RACH but carefully cause can make UL interference.
3. check / change transmit power of CPICH to low level is use to min coverage but max performance ex 33 dBm to 30 dBm
4. check overload control due to power set transmit and receive power of cell (Ptx and Prx)

this is just clue to more focus on understanding your problem..
read your document related to 4 item above for their setting and their function related to RRC Conn Reject because Admision control not admited

CMIIW
BR//edy

Thanks Edy. I have qrxlevmin set to -115 and qqualmin set to -24. This is the standard setting for IFHO_C(Core site). I can of course reduce this to a lower value.
CPICH power is set to 312.
Which parameter do I change for RACH? aichpower?
Ptx and Prx..is that with E*******? I thought it was with Nokia.

Thanks.

[zhanglw268]

The statistics looks very strange as pmTotUtranRejRrcConnReq is zero and pmnoFailedafterAdm isn't,

Can you compare the statistics pmTotNoRrcConnectReq and pmTotNoRrcConnectSetup except
pmTotNoRrcConnectReq
pmTotNoRrcConnectReqSuc
pmTotUtranRejRrcConnReq
pmnoFailedafterAdm

Regards

David

Thanks David. Yes the cells is in the RNC border and there is pilot pollution. Tilts have been changed but still there is accessibility issues. I know there would be a lot of signalling that would also cause the accessibility failures.

[tsiz]

The statistics looks very strange as pmTotUtranRejRrcConnReq is zero and pmnoFailedafterAdm isn't,

Can you compare the statistics pmTotNoRrcConnectReq and pmTotNoRrcConnectSetup except
pmTotNoRrcConnectReq
pmTotNoRrcConnectReqSuc
pmTotUtranRejRrcConnReq
pmnoFailedafterAdm

Regards

David

Hi David,

here is the info you requested. I have some more counters added for the ROP.

pmTotNoRrcConnectReq 2414
pmTotNoRrcConnectReqCs 152
pmTotNoRrcConnectReqCsSucc 54
pmTotNoRrcConnectReqSuccess 860
pmTotNoRrcConnectSetup 952
pmTotNoRrcReq 2504
pmTotNoTermRrcConnectReq 407
pmTotNoTermRrcConnectReqCs 64
pmTotNoTermRrcConnectReqCsSucc 21
pmTotNoTermRrcConnectReqPs 125
pmTotNoTermRrcConnectReqPsSucc 46
pmTotNoTermRrcConnectReqSucc 138
pmTotNoUtranRejRrcConnReq 0
pmTotalTimeDlCellCong 0
pmTotalTimeHsdschOverload 0
pmTotalTimeUlCellCong 0

[zhanglw268]

From the statistics I can see that there are a huge difference between pmTotNoRrcConnectSetup and pmTotNoRrcConnectReq, but pmTotNoUtranRejRrcConnReq = 0, I can see 2 possible reasons:
1. Current cell doesn't have dominant pilot CPICP, UE sent "RRC Connection Request", then left the current cell after T300 times out, and reselected
to another cell in the same RNC, UE sent "RRC Connection Request" in another cell, RNC will not peg the pmTotNoRrcConnectReq again in other cells so current cell's pmTotNoRrcConnectSetup is less than pmTotNoRrcConnectReq as there is no RRC Connection Rejection, you can check other cells statistics to see whether it is true, otherwise
2. software bug that prevents RNC to send RRC connect Setup in the current cell or neighbour cells, also RNC prevents pmTotNoRrcConnectReq pegging.

To narrow down the problem, you can do the following:
a). comparing current cell's T300 setting, then increase T300 to see whether the situation improves
b).check current cell's neighbours to see whether they have RRC setup success rate exceeding 100%
c). change Qrxlevmin to -110 dBm or higher and Qqualmin to -18 dB or higher, this will reduce the coverage area so UEs from surrounding cells will
reselect to this cell with good quality coverage
d). reduce the CPICP to reduce the current cell coverage so UEs will reduce the cell reselect to this cell at the cell edge
e). Check current cell's neighbours, reduce neighbour cells' CPICP so the current cell CPICP becomes dominant, UEs can stay in current cell to make calls
f). make the cell radius large to accept calls
g). reset the NodeB in maintenance window to see whether the problem disappears then comes back, if not possible, you need to raise a trouble ticket with E******* support.

Regards

David


Hi David,

here is the info you requested. I have some more counters added for the ROP.

pmTotNoRrcConnectReq 2414
pmTotNoRrcConnectReqCs 152
pmTotNoRrcConnectReqCsSucc 54
pmTotNoRrcConnectReqSuccess 860
pmTotNoRrcConnectSetup 952
pmTotNoRrcReq 2504
pmTotNoTermRrcConnectReq 407
pmTotNoTermRrcConnectReqCs 64
pmTotNoTermRrcConnectReqCsSucc 21
pmTotNoTermRrcConnectReqPs 125
pmTotNoTermRrcConnectReqPsSucc 46
pmTotNoTermRrcConnectReqSucc 138
pmTotNoUtranRejRrcConnReq 0
pmTotalTimeDlCellCong 0
pmTotalTimeHsdschOverload 0
pmTotalTimeUlCellCong 0