cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
726
Views
5
Helpful
2
Replies

Discrepancy between compand type and codec: what happens?

raziel78kain
Level 2
Level 2

Hello,

we have a CME with a CUE and some BRIs.

Since we are in Europe, we have configured the "compand-type a-law" under every "voice-port" related to these BRIs (to avoid the distorted audio).

But, since the CUE only supports the G.711 u-law, we have configured the "codec g711ulaw" in every "dial-peer" pointing to the CUE.

We want to point out that we don't have transcoders and/or MTPs configured on the CME.

So, our question are the following:

  • What happens exactly due to this discrepancy between the compand type and the codec?
  • Do we have a double conversion for the call (voice -> a-law -> u-law)? If yes, who makes this conversion? Is it the DSP on the PVDM?
  • Do we have a loss of quality for the call due to this discrepancy?

TIA and regards.

2 Replies 2

Suresh Hudda
VIP Alumni
VIP Alumni
  • What happens exactly due to this discrepancy between the compand type and the codec? ------) Nothing, pvdm(dsp) will take care of it, you don't need to configure transcoder or mtp for it, you require them when you have two voip call legs on different codec. Here one leg is pstn and another one is voip so dsp on pvdm will take care of this conversion.
  • Do we have a double conversion for the call (voice -> a-law -> u-law)? If yes, who makes this conversion? Is it the DSP on the PVDM? -----) Yes, it's dsp on pvdm.
  • Do we have a loss of quality for the call due to this discrepancy? ----) I don't think so.

Suresh

Hello Suresh,

are you sure about the absent loss of quality?

Because meanwhile I've found this document:

https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-G.711-198811-I!!PDF-E&type=items

where it's stated that:

"... in most of the analogue voice frequency signal range the additional quantizing distortion caused by µ-A-µ or A-µ-A conversion is considerably lower than that caused by either µ-A or A-µ conversion".

So, the A-µ conversion (as in our scenario) causes an additional quantizing distortion.

I wonder how much loss of quality is caused by this additional distortion, for example, in terms of MOS...

Anyone can help me to better understand this thing, please?

Thanks and regards.