Dear Experts.
I have getting MOS under maximum 3 - no value is above 3.
What can be the issue?
Please share your suggestion.
And how to analyze low MOS from log file (Drive Test)?
In whole drive test AMR codec is 4.75 and frame rate is N. 12.20kb and /sec.
You might have the voice sample set @ WB-AMR but your network is only NB_AMR.
We have NB 12.20 Kbps.
Check the script on the drive test. the wave you use must not be adequate.
You should use a wave that is also NB + the algo on the drive test tool must be specified to used NB.
What tool are you using to to the drive test?
I think the issue is with the configuration of the drive test tool.
Tool is Huawei Mobile PHU (GENEX Probe Handset Unit).
Check the ALGO that you are using.
You need to have POLQA and not POLQA-WB.
But i do not have experience with Huawei Drive Test…
For my Drive Test tool these are the option i have:
Please provide a simple table showing the summary of codec assigned during this MOS test.
If your MOS samples were on NB-AMR4.75 then your MOS will be lower than 3.
And why the network did not assign below codecs? Are they configured?
Please check EcNo vs MOS and the codec rates.
Please be sure that the device is capable with all Codecs supported by the network.
AMR FR 12.2 you should get until 4.10 max value.
Is from MB to MB (Mobile to Mobile) Measurements? Or MB to Fixed?
In addition to above points:
-Some times if you are calling predefined test number and not a normal user number, it has limited codec rate.
Try normal sim cards in MOC & MTC if this is not the case
-Make sure codec set rates are defined correctly from RAN & Core sides
-Check transition threshold parameters between codec rates are defined correctly
-Try the test under the site in perfect radio conditions
-After you finalize this, try enable WB-AMR for better MOS
In addition to above points ,
In Huawei , there are many features to enhance voice quality and so on MOS ex:
Hello All.
What are POLQA files options?
Select the file type according to the measured voice codec rate if using EVS swb for example you need to select polqa SWB audio file and so on.
For OTT make sure to select SWB or FB as most of the OTT application are running on opus swb/fb.
Thanks, but i would like to know the scenario of algorithms intervals and messages.
Message will be synthesized for swb and fb and nb based on the home country’s spoken tongue (available in English and Japanese…etc ).
So it depends coming to algorithm some I insights are described on the ITU-T SG 12 for itu users.