- This topic has 9 replies, 1 voice, and was last updated 17 years, 2 months ago by jake.
14th August 2003 at 21:32 #36319OlguetaGuest
Why would a mobile not be able to decode the BSIC? I think there are two reasons: 1. Because of interference. 2. Because the MS is not able to get the SCH channel and read the BSIC information. Am I right?
We are experiencing the second situation. How often should the MS try to measure a SCH? From looking at TEMS log files we see mobiles not measuring SCHs for minutes and reporting Synchronization Errors. In cases where the MS does read the SCH we looked at the frame offset and quarter bit offset of a neighboring cell with respect to the serving cell and there was a significant drift. So, what would be causing synchronization problems?
Thanks17th August 2003 at 04:35 #36320jjohnwongtpGuest
I read about your questions, MS need to decode BSIC of serving cell and Neighbor Cell due to camping on a new cell need this synchronization, otherwise, MS will know about to sync with the SCH;
The problem of MS can not get the Sync channel sometime, then, it will be several situtation you need to check :
1. If you use an Agilent Gsm/Gprs radio set to test with your MS, can it sync with its SCH ?
2. If you can sync with the above situtation, but MS is not able to sync with some SCH with some network, then, it need to check with several factor, is it a serving cell of BSIC that MS can not sync with or Neighbor cell of BSIC ?18th August 2003 at 05:41 #36321YamamotoGuest
I encountered similar problem as yours. Also looking for solution.
I tried something more: I forced the TEMS to explicitly select the cell with SCH problem. It could camp on the cell and worked fine, but failed to decode the SCH contents of all the neighbours. So, the phenomenon is, the MS can either decode the SCH of the cell being tested, or the SCH of all the neighbours. Is it the same on your side?18th August 2003 at 20:11 #36322OlguetaGuest
What we used for testing was TEMS Investigation 4.0 with two phones, a T62u and a R520m. I should review the logs file but I think that the MS could decode the SCH from the serving site but not the SCH from the neighbors.
Did you check the frame offset and quarter bit offset? Do you know what’s the acceptable drift?
What’s the synchronization source of your network, do the Base Stations get their own clock from a GPS source or from the T1’s?
In our case we use the T1 as synchronization source, and currently we have a T1 expert investigating this issue.
Are you working with Ericsson BTS or Nokia?19th August 2003 at 09:39 #36323jjognwongtpGuest
As I know about the MS from its tracing, MS needs to decode serving cell and several strongest neighbor cells, this is very important preparation for dedicated link, otherwise, MS must do the SCH sync again when the MS go into Cell reselection of new cell among the neighbors;
so I check with spec. it needs to implement this feature
(A), “A procedure shall be implemented in the MS by which it monitors the downlink RX signal level and quality from its serving cell and the downlink RX signal level and BSIC of surrounding BTS.”,
(B), “The MS shall attempt to check the BSIC for each of the 6 strongest non?serving cell BCCH carriers at least every 30 seconds, to confirm that it is monitoring the same cell. If a change of BSIC is detected then the carrier shall be treated as a new carrier and the BCCH data re-determined.”
So if the MS can not do this, this means the MS have some issue not meet with spec.;19th August 2003 at 16:00 #36324OlguetaGuest
The MS do meet the specs, meaning that they attempt to listen to the timeslot 0 on the BCCH carrier looking for the SCH and decode the BSIC, but they fail. We know that because on the logs files we can see “Synchronization error”. There must be some kind of frame synchronization error that doesn’t allow the MS to decode the information on the SCH.
It doesn’t happen all the time and it seems that the Nokia 6200 are more sensitive to the problem.20th August 2003 at 04:13 #36325YamamotoGuest
The fact is, when the MS can successfully camp on the cell, either the test cell or the neighbours, the frame offset and quarter-bit offset are of similar range. However for those cells that the MS have problem in decoding the SCH, both offsets are zero.
We just have the BTS sync to the E1 link, no GPS. We use Nokia BTS.
BTW, do anybody have more info about the frame offset and quarter-bit offset? Which ETSI doc should I look up?20th August 2003 at 10:28 #36326jjohnwongtpGuest
Hi Olgueta and Yamamoto,
Thanks for your response;
I understood your description and know how to clarify this issue by the approach below :
(1). Does the BSIC decoding issue only sensitive to some MS and some location of field coverage ?
(2). Do you have some trace relative to the SNR/Frequency error of MS decoding of Neighbor Cell ?
In the Spec., the MS need to decode the BSIC for each of the 6 strongest non?serving cell BCCH carriers at least every 30 seconds, so any kind of the frame offset and quarter-bit offset situation also needs to include, this means it could happen relative to SNR, Frequency error and signal level !!!
Because we believe the Nokia MS is qualified.21st August 2003 at 14:42 #36327OlguetaGuest
The problem affects all mobiles but it seems that the Nokia 6200 is more likely to experience it.
A typical situation when the BSIC decoding would fail is when there is interference or when the signal level is too low. In our case none of this is happening, the MS fails decoding the BSIC of a neighbor even though the signal level from that neighbor is -75 dBm.28th August 2003 at 09:52 #36328jakeGuest
MS takes it’s reference frequency from the camped cell. In case there’s a differency in synchronisation of two BTSs taken from PCM or in frequency multiplication in BTSE (to rf) then the two BTS are not frequencywise synchronised and BSIC decoding gets hard or impossible. While MS is camped on a BTS with abnormal reference it can still easily detect (decode BSICs) adjacent cells in the same BTS but not the once in other BTSs.