Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors

The SDCCH assignment fail from the LU.

Viewing 15 posts - 16 through 30 (of 37 total)
  • Author
    Posts
  • #51169
    pix
    Guest

    what is the value of sdcch assignment failure rate due to radio problem ?
    85% ??

    if Abis or Ater link is over satellite, this might explain the high failure rate : you need to set longer values for several timers (for gsm and gprs).

    Can you list the values of the following QoS indicators, per cell :
    sdcch drop rate
    tch assignment failure rate
    tch allocation efficiency (per TRX)
    tch drop rate

    RACH, PCH, AGCH loads & numbers

    SDCCH traffic model (rate of LU, OC, TC, IMSI detach, etc)

    SDCCH traffic BH per TRX
    TCH traffic BH per TRX

    Radio Measurements, per TRX:
    – RXLEV DL AVG
    – RXLEV UL AVG
    – Path Balance Avg
    – Timing Advance Avg
    (these indicators are at the bottom of the list of indicators in RNO)

    The path balance might be the problem… Please list all these values, for one typical day.

    Regards,
    Pix

    #51170
    Spit
    Guest

    Hi everyone.
    pix, can you explain a bit the concept of path balance? and the usual problems that reflects a bad value of path balance?
    thx =)

    #51171
    pix
    Guest

    hello spit,

    path balance is entirely due to a faulty hardware, it shows a higher loss in the BTS’s reception path, or on the opposite, a higher loss in the BTS’s transmission path.

    Depending on your formula :
    path bal > +5dB : problem in the Tx
    path bal < -5dB : problem in the Rx Most common problems are : faulty TRX or faulty combiner Other possible problems : faulty feeders, connectors, jumpers, antennas And finally, it might also help you find out crossed feeders (Rx on one sector, and Tx on another sector)

    #51172
    sunny
    Guest

    Hello:
    What is the refrence point for measuring the path balance ?
    Where will be get those measuremnet ? how will the power control algorithim effect those values?

    #51173
    pix
    Guest

    reference point ? you mean where it is measured ? It is measured at the TRX connector.

    you can get the path balance from Abis Measurement Results, or from radio measurement statistics, in alcatel systems (= RMS)

    the formula is :
    Path balance = (RxLev UL – MS Tx Pwr) – (RxLev DL – BTS Tx Pwr)

    You can obtain these four values from a measurement result. By averaging all results on one hour, you can get a prety good idea of the path balance in each TRX.

    Power control is using the same four values, so both path balance and power control will show the same effects :
    If path balance = 0, the power control will attenuate the DL Tx Pwr and the UL Tx Pwr by the same amount.

    But power control is not a side effect of path balance ! They are just using the same inputs, so obviously they’ll reflect the same logic.

    Regards,
    Pix

    #51174
    sunny
    Guest

    We have been facing issues with the uplink Signal strength and qulaity in 850/1900 dual band Ericsson system. The problem is both at the setup as well as in dedicated mode resutling at severe sdcch failure and TCH drop call. The problem is more accute on 850
    The TA data indicated that some of the cells are getting the samples from unplanned coverage areas (Low Siganl Erlang) from new developments

    The following have been already tried

    1Dual Band TMAs(The drop shifted from uplink to drop due to both link)

    2Frequency Retune

    3Load Balancing (1900/850);
    4 Neighbor scrubs
    5 scrubbed Hardware issues

    We can’t change Accmin, power or downtilt as it will affect the customer coverge.

    Is there a way that we can limit the overshooting cells without accmin/power or downtiling.

    Thank for you help

    #51175
    Alexx
    Guest

    Dear Pix,
    SDCCH assignment failure is about 10-15% SDCCH assignment failure = (MC149/(MC148+MC04)) but MC04(SDCCh bloking_congestion)=0).
    We use only terrestrial Abis and Aters.
    For the one typical bad cell (21.03.08 – from RNO):
    1) SDCCH Drop Rate = 0% (MC138+MC137+MC07)=0;
    2) TCH Assignment Fail=0% and TCH Successful Rate = 99% (MC703/MC718);
    3)TCH Drop Rate = 1.4%;
    4)SDCCH traffic BH per TRX = 0.17 Erl (SDC only on BCCH-tre);
    5) TCH traffic BH per TRX = 2.95(tch-tre)+ 0.72 (bcch-tre);
    RMS:
    – RXLEV DL AVG= – 77.61 dBm
    – RXLEV UL AVG= – 81.22 dBm
    – Path Balance Avg=17.65
    – Timing Advance Avg=4..8
    About MC149 – for the mentioned cell this value is about 50-60 failures per hour. T3101= 3 sec (maybe we have to change this period (make it longer?)
    The value of Path balance is really high and it’s very strange, especially because we use TMA for 1800 cell !!!
    I know that there was a Fault report in Alcatel “FR A45/131698” (MS Path Balance values are incorrect for DCS 1800 cells) but it had to be corrected in B8 (we use BSS B9 MR6).
    Thanks a lot for your assistance!!

    #51176
    pix
    Guest

    hi,

    thanks for all these values, that’s great.

    your path balance has a very very bad value (what about path balance value per TRX ?), it shows that your TMA is working too good : your coverage is strongly downlink limited. You might have a problem on your Tx path (your Rx is 17dB better than your Tx)

    Therefore, a MS might be able to send a Channel Request on the RACH (UL), but it is unable to hear the Immediate Assignment. It is indeed a readio problem, a real one. Changing the timer T3101 will not help. (you can check with drive tests).

    the “1800 MS” problem is not here anymore, as far as i know, i’ll check it out.

    your SDCCH assignment failure rate is a bit high (10 to 15% is not critical, but above 10% in an urban area might indicate a problem), yet your overall QoS is fine.

    50 to 60 SDCCH assign failures per hour is an absolute number, you got to divide it by the number of SDCCH requests (MC04+MC148 = about 600)

    Next step : check if path balance is on both TRXs, or just one.
    try locking the BCCH TRX, in order to relocate the BCCH on the second TRX, if it has a good path bal.

    Regards,
    Pix

    #51177
    Alexx
    Guest

    Dear Pix,
    I’ve locked TCH-tre in order to check Path Balance on BCCH-tre (there were no samples for it before). This value was the same ~ 15..16.
    But we have found the problem for this cell : ))) It was a faulty equipment (like GSM-modem, security system of a large store). Every ~ 30 it sended a Channel Request message with cause ” Originating Call …” but after receiving Immediate_Assignment message didn’t do anything – that’s why we have abnormal MC149 on this cell. We got very clear picture in Abis trace 🙂
    But it’s only one of our problem cells. And about average Path Balance for our 1800 cells – it’s really ~ 14..16 !! I think it means that we have serious problem in TX.

    #51178
    pix
    Guest

    hello alex,

    wow, good work :)) so the external equipment generates a signal that looks like a Channel Request ? Amazing… However, I don’t think it impacts the QoS for the subscriber.. it’s just an optimizer challenge to fix this kind of issues 🙂 (and don’t we love it ??)

    i assume you are in B7 ? there was a fault report about a delta of 13dB between the real value and the measured value of path balance (you can check the RMS Tx MS Power, the problem is there)

    It is now corrected since a long time, but maybe you’re still using an old version of B7 ? You might want to escalate this problem to your alcatel project manager or local RTAC, in order to get an updated release (?)

    #51179
    Alexx
    Guest

    Hello Pix,
    we use B9 mr6.4, but obviously I read old docs :((
    Unfortunately I’ve chosen for research bad example (about this strange cell :). But analizing the distribution of Path Balance for the whole 1800 range has given unexpected result – average value ~14..15.

    #51180
    Pix
    Guest

    oh yes, you said it earlier : B9 MR6. you’re quite up-to-date !! Which country is it ? (if you don’t mind)

    All your 1800 cells have this kind of path balance ? I mean, this is a REALLY bad path balance, the call drop rate, ho success, call setup success, and most KPI indicators have to be terribly bad !

    Since i doubt it is a real hardware problem, i think that’s a RMS issue : could you raise this issue to your RTAC ? They’ll probably give you a patch to fix it.

    #51181
    Alexx
    Guest

    Dear Pix,
    I’m from Ukraine. I work in UMC “Ukrainian Mobile Communication”. Alcatel BSS is used only in 3 region, but it’s ~ 14000 of cells. I’m an engineer in group of Network and service quality.

    #51182
    ALexx
    Guest

    Dear Pix,
    About our KPIs:
    1)we have no problems with Drop Call Rate – it’s 0.8%;
    2)but Call Setup Successful Rate (CSSR) is only 97,3%. Moreover, we have inadmissible high rate of cell with CSSR is less then 95% (mainly due to low Immediate Assignment Rate (Successful or Unsuccessful SDCCH assignment rate)) it’s about 34% of cells.
    We have analyzed (and partly solved) several typical problems: Phantom RACH, Interference on BCCH, different HW problems, and have stopped at problem of Unbalanced Power Budget which is directly connected with Path Balance. Thank you very much for your advice – I’ve just sent detailed description of this problem to our TAC,I hope they will help us.
    Thanks for your assistance.

    #51183
    Emmanuel Aboagye Domfeh
    Guest

    Hi,

    I am working in Millicom Ghana as an Optimizer and have this precurrent condition of Rx diversity. I have traced all cables but seen no swap of cables.
    Could you help me out with more diagnostic procedures to follow?

Viewing 15 posts - 16 through 30 (of 37 total)
  • The forum ‘Telecom Design’ is closed to new topics and replies.