AMR FR : Huawei solution

[gogotchiya83]

Dear Experts,
We have a problem in our MOS in compared with our competitor, last time we made a DT and we find some strange result, exampl : AMR FR 12.2 Kbps with MOS value less than 1.5 !!!


kHRIBGA.JPG











When we discover that we decided to disable the AMR FR (test made on city: good Rxlev with good Rxqual) and the strangly the MOS become better
How you can explain that ?

[khurrambilal01]

If Intra-Cell Handover is enabled in the network then what is happening is that for bad quality samples BSC is giving a Intra-Cell Handover command to shift that call to AMR FR (To improve user perception) but the problem is if quality was bad on one of the timeslot there are bright chances that it would be bad on other timeslot as well (Even on other TRX),
Once you disabled the AMR FR then no Intracell HO was taking place and the bad quality call was being Handed over to neighor cell which has more probability of having good quality as compared with last serving cell's intra timeslots/TRXs.

So, this indirectly justifies your MOS result (If you have intra cell HO enabled in your network before disabling AMR FR)

[gogotchiya83]

Tkx for your Nice answer, can you share some docs treating this issue?

[gogotchiya83]

Dear experts, Still waiting for you help? As you can see bellow in some area we have bad MOS even with the use of AMR FR 12.2 in good radio condition (C/I > 30)
How you can explain that?
ANOMLAIE.jpg

[beniy]

Dear experts, Still waiting for you help? As you can see bellow in some area we have bad MOS even with the use of AMR FR 12.2 in good radio condition (C/I > 30)
How you can explain that?
ANOMLAIE.jpg

My friend,
I think Iknow what is problem,
Just tell me it happened in which BTS type?
you are using A OVER IP or not in your network?

[gogotchiya83]

thanks, i hope that you will be able to help us (for the momemnt we decided to deactivate the AMR FR, and we see a real improvement in our MOS trough the DT and also in the VQI, so we are sure that the problem is comming from the AMR FR).
We use BSC 6900, A over IP...
Any help?

[beniy]

thanks, i hope that you will be able to help us (for the momemnt we decided to deactivate the AMR FR, and we see a real improvement in our MOS trough the DT and also in the VQI, so we are sure that the problem is comming from the AMR FR).
We use BSC 6900, A over IP...
Any help?

My friend,
What is BTS type? this is very important...Mostly problems refer to BTS 312 and some Codec.
ANd please tell me which equipment you done MOS test and analysis.

Answer me we had complete analysis about that and we found the issues

[gogotchiya83]

Tkx beniy for your answer,

We have BTS 3900, we used NEMO for the DT with mobile supporting AMR...
We did many test and we seen an improvement of the MOS after disabling the AMR FR (test done with NEMO, PROBE)...
So no we decided to deactivate this feature in the whole network...

Any help PLZ

[beniy]

We had exact problem please send me your email to send you something.
And please send me MOS TEST before and after deactivating FR AMR.
And tell me you have Just BTS3900? BTS Type is very important.
that's long story There are diffrent problems and I need some statistic to tell you exact problem in your network..
Check your PM

[airian]

We had exact problem please send me your email to send you something.
And please send me MOS TEST before and after deactivating FR AMR.
And tell me you have Just BTS3900? BTS Type is very important.
that's long story There are diffrent problems and I need some statistic to tell you exact problem in your network..
Check your PM

Dear Beniy do share with us too about your analysis plz. will increase our knowledge too.

Dear Gogotchiya we also have an A over IP network of Huawei with BTS3900 and we had done Benchmark with NSN and E******* non A over IP with nemo. the results were great with Huawei AMR. You should involve your vendor in this scenario.

Also plz share your settings for MOS calculation i.e. audio sample size. You can also verify your MOS equipment as MOS values can be very low even with weak battery backup of MOS equipment, wrong loudness settings on MS etc.