- This topic has 36 replies, 1 voice, and was last updated 12 years, 6 months ago by m@u.
25th February 2008 at 04:25 #51154SKYGuest
We want to reduce the SDCCH Assignment fail in our network. After we took a look on the cause of assignment fail,
we found that the most was come from the Location update.
We found in the protocol message that the failure always occurred after the AUTREQ and CICMD.
Could you please tell me that what is the cause of SDCCH failure from the Location update?
Thank you very much in advance. 🙂25th February 2008 at 09:39 #51155PixGuest
I’m surprised you can say the problem occurs only during Location Update : during a Call Setup, the SDCCH phase is almost the same (Ciphering & authentication)
The problem you are describing is not a SDCCH assignment failure, but it is a SDCCH drop. So it is probably linked to a radio problem, such as interference or bad coverage, which occurs on the TRX that is carrying the SDCCH timeslot.
If the problem is only on this TRX and not on the others, then it is probably a faulty TRX.27th February 2008 at 19:33 #51156ZbigniewGuest
Hi Sky, what is your default value for timer T200 ? It sounds like a Layer 2 problem. Greetings Z.28th February 2008 at 07:20 #51157SKYGuest
Thanks for your recommend.
We are using the T200=300ms corresponds to Abis terrestrial link. (in Alcatel have only 2 choices, Terrestrial=300ms and Satellite=1000ms.)
More information and more question about this problem.
We found the time after the CLUPD(send from BSC to MSC) before the CLCMD(send from the MSC to the BSC has the cause value=call control) is equal about 11s.
From this it look like the time out behavier.
Could anyone please tell me that which timer should it be?28th February 2008 at 07:43 #51158PixGuest
i don’t really want to disturb anymore… 🙂28th February 2008 at 08:27 #51159SKYGuest
Why not, Pix??? Please!!! 🙂28th February 2008 at 09:48 #51160ZbigniewGuest
Hi Sky, I recommend to set the timer T200 to 1.000 ms. We noticed the same problem in Nokia and Siemens BSS. Greetings Z.28th February 2008 at 10:00 #51161NarChaiGuest
it might be “Phantom RACH”29th February 2008 at 02:43 #51162SKYGuest
We found that the 11-12s time out is the T3109 which is related to the Radio link timeout.
So it probably linked to the radio problem like Pix said.
We are investigating and finding out the solution.
Any recommend are welcome!!! 🙂
BRs,8th March 2008 at 16:47 #51163DariushGuest
Can You tell me is this site border site? if yes please check the bsc borders and try to decrease coverage for this site to have less LU In your networks.9th March 2008 at 03:43 #51164sunnyGuest
Try increase the CRH( Cell reselection Hysteris) on the border cells . This will decreasse the ping pong around the LA Border and so the sdcch faiures will decrease14th March 2008 at 17:51 #51165Zaki RyneGuest
I am working as a network performance team Engineer. During drive test we have 2 cell phones adjusted which call one another(master & slave phone) continuously. But when they crosses the location border, the call drops.
Can anybody give me a suggestion what actually happens in the border? Why do they always drop?14th March 2008 at 21:16 #51166PixGuest
Have you looked at the messages logged in the drive test (layer 3)
– what happens just before the drop ?
– what about qoS statistics in serving and target cells ? do you notice something in terms of handover failures or call drops ?
i would suspect that the two location areas are located in different MSC’s, right ?
you could check neighbour definition and LAC definition in the different MSCs (can’t tell you more than this, you must check with NSS engineers)
you can also ensure that both cells are using the same ciphering algorithms, as it can lead to a drop when switching ciphering in dedicated mode (as I experienced, but it might not be true with every vendor !)
if the cells are from same MSC, but different BSC, then there is probably something you need to investigate on the A interface.
Altogether, you can investigate further with detailed QoS indicators and/or A interface traces, and simply have a look to your layer 3 messages in the drive test.
pix16th March 2008 at 16:34 #51167Zaki RyneGuest
Thanks for the suggestion. I will try them out.
Zaki20th March 2008 at 19:34 #51168AlexxGuest
I work with BSS Alcatel. We have a serious problem with Immediate Assignment: for 20% of cells it`s ~ 85-90%.The problem connects to counter MC149 – expired T3101 (3 sec), i.e. within 3 sec MS doesn`t acknowledge occupation of SDCCH. We have changed BCCH/BSIC to chek the version with Phantom RACH and BCCH-interference, have changed all equipment in the cell (trxs & ANc) but – without any results 🙁
Do anybody have similar problem?