FPGA Central

Hi all,

I am working on acoustic echo canceler for mobile telephony. I have tw
doubts to clarify:

1) Avi perry in his book "Voice Quality Engineering in Wireless Networks
has mentioned that although speech signal power is distributed almos
uniformly across the entire spectrum, the acoustic echo power concentrate
most of its energy in the 1700-2500 Hz range. Anybody having any idea o
the possible explanation.

2) I am searching a bulk delay estimation method (for AEC)that can be use
in the presence of heavy non-linear distortions. These non linea
distortions are introduced by : GSM codec, Speaker, Mechanical/Acousti
coupling. I will be very thankful if anyone can guide me through.

Any good books that deal with Acoustic Echo Cancelers in wireles
network.

Thanks
Sankalp

createdon2003 wrote:
> Hi all,
>
> I am working on acoustic echo canceler for mobile telephony. I have two
> doubts to clarify:
>
> 1) Avi perry in his book "Voice Quality Engineering in Wireless Networks"
> has mentioned that although speech signal power is distributed almost
> uniformly across the entire spectrum,

????

Average speech signal power has maximum at few hundred Hz and then rolls
down towards the high frequencies with the rate about 10dB/oct.

> the acoustic echo power concentrates
> most of its energy in the 1700-2500 Hz range. Anybody having any idea of
> the possible explanation.

A misprint, perhaps. More likely, they meant 170...2500.

> 2) I am searching a bulk delay estimation method (for AEC)that can be used
> in the presence of heavy non-linear distortions.

Run a traditional *.LMS algorithm over all span of the expected delay.
Then look at the resulting impulse response; the bulk delay part will be
seen clearly. To keep the computations reasonable, you can decimate the
signal to the lower sampling rate.

> These non linear
> distortions are introduced by : GSM codec, Speaker, Mechanical/Acoustic
> coupling. I will be very thankful if anyone can guide me through.

Auch. I doubt you can achieve any good result if you have a vocoder in
the EC loop.
The compensation of the small nonlinearities is feasible; it is better
done in the frequency domain EC. However I don't know of any practical
EC system which employs the nonlinear compensation.

> Any good books that deal with Acoustic Echo Cancelers in wireless
> network.

The EC loop is closed at the ends of the link; it doesn't matter what
kind of network is in between.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

thanks for the reply...
>
>
>createdon2003 wrote:
>> Hi all,
>>
>> I am working on acoustic echo canceler for mobile telephony. I hav
two
>> doubts to clarify:
>>
>> 1) Avi perry in his book "Voice Quality Engineering in Wireles
Networks"
>> has mentioned that although speech signal power is distributed almost
>> uniformly across the entire spectrum,
>
>????
>
>Average speech signal power has maximum at few hundred Hz and then roll

>down towards the high frequencies with the rate about 10dB/oct.
>
>> the acoustic echo power concentrates
>> most of its energy in the 1700-2500 Hz range. Anybody having any ide
of
>> the possible explanation.
>
>A misprint, perhaps. More likely, they meant 170...2500.

Actually I have verified it through a real recording, but couln't find th
reason. I am suspecting the mechanical coupling because of the poo
casing..ne idea???..how does that coupling is modeled (Frequency response
n all)..couln't find it anywhere..

>
>> 2) I am searching a bulk delay estimation method (for AEC)that can b
used
>> in the presence of heavy non-linear distortions.
>
>Run a traditional *.LMS algorithm over all span of the expected delay.
>Then look at the resulting impulse response; the bulk delay part will b

>seen clearly. To keep the computations reasonable, you can decimate the
>signal to the lower sampling rate.
>
>> These non linear
>> distortions are introduced by : GSM codec, Speaker
Mechanical/Acoustic
>> coupling. I will be very thankful if anyone can guide me through.
>
>Auch. I doubt you can achieve any good result if you have a vocoder in
>the EC loop.
>The compensation of the small nonlinearities is feasible; it is better
>done in the frequency domain EC. However I don't know of any practical
>EC system which employs the nonlinear compensation.
>
>> Any good books that deal with Acoustic Echo Cancelers in wireless
>> network.
>
>The EC loop is closed at the ends of the link; it doesn't matter what
>kind of network is in between.

Sorry probably it wasn't very clearly written. I meant GSM networ
(Vocoders are present in both directions).
>
>
>Vladimir Vassilevsky
>DSP and Mixed Signal Design Consultant
>http://www.abvolt.com
>
>
>
Sankalp

createdon2003 wrote:


> Sorry probably it wasn't very clearly written. I meant GSM network
> (Vocoders are present in both directions).

If you have vocoders in the loop, the EC is not feasible.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

>> the acoustic echo power concentrates
>> most of its energy in the 1700-2500 Hz range. Anybody having any idea
of
>> the possible explanation.
>
>A misprint, perhaps. More likely, they meant 170...2500.

Actually I have verified it through a recording (not sure whether speake
phone was on/off)that acoustic echo's power was present mainly betwee
1700-2500Hz, but couln't find the reason. I am suspecting the poor acousti
isolation due to cheap casing..any idea???..how does that coupling i
modeled (Frequency response n all)..couln't find it anywhere..

Sankalp

On Sep 3, 1:28*am, "createdon2003" <createdon2...@yahoo.com> wrote:
> >> the acoustic echo power concentrates
> >> most of its energy in the *1700-2500 Hz range. Anybody having any idea
> of
> >> the possible explanation.
>
> >A misprint, perhaps. More likely, they meant 170...2500.
>
> Actually I have verified it through a recording (not sure whether speaker
> phone was on/off)that acoustic echo's power was present mainly between
> 1700-2500Hz, but couln't find the reason. I am suspecting the poor acoustic
> isolation due to cheap casing..any idea???..how does that coupling is
> modeled (Frequency response n all)..couln't find it anywhere..
>
> Sankalp

You need to explain the system that you are trying to apply this to
and at what point in the system you are planning on putting aoustic
echo cancelling. Ex: Maybe you have a talker in a car or a room where
the audio (original and echoes) are going into a system that is doing
bandwidth compression on the front end. If you put the echo
cancellation before the coder, then the distortion of the coder will
have little effect.

In other situations the distortion could have a great effect. So,
explain please, in some detail.

Dirk Bell
DSP Consultant

Dirk Bell wrote:

> On Sep 3, 1:28 am, "createdon2003" <createdon2...@yahoo.com> wrote:
>
>>>>the acoustic echo power concentrates
>>>>most of its energy in the 1700-2500 Hz range. Anybody having any idea
>>
>>of
>>
>>>>the possible explanation.
>>
>>>A misprint, perhaps. More likely, they meant 170...2500.
>>
>>Actually I have verified it through a recording (not sure whether speaker
>>phone was on/off)that acoustic echo's power was present mainly between
>>1700-2500Hz, but couln't find the reason. I am suspecting the poor acoustic
>>isolation due to cheap casing..any idea???..how does that coupling is
>>modeled (Frequency response n all)..couln't find it anywhere..
>>
>>Sankalp
>
>
> You need to explain the system that you are trying to apply this to
> and at what point in the system you are planning on putting aoustic
> echo cancelling. Ex: Maybe you have a talker in a car or a room where
> the audio (original and echoes) are going into a system that is doing
> bandwidth compression on the front end. If you put the echo
> cancellation before the coder, then the distortion of the coder will
> have little effect.
>
> In other situations the distortion could have a great effect. So,
> explain please, in some detail.

He is talking about the local feedback cancellation in the hands-free
type of thing, and his main concern is not the acoustic echo per se, but
the direct coupling between the speaker and the mike due to the
deficiencies of the housing. If the speaker and the mike are arranged
poorly, there is only so much that you can improve by the DSP.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

may be it will be more clear by a diagram.

--far_end-------------------|ENC|-------|DEC|--------//
| | speaker
| \\
AEC
| LEM wit
T_60~60ms
|
--to far end----------------|DEC|-------|ENC|--------O microphone
|
AMR codecs are to be used for ENC_DEC, I am simulating the worst cas
MR475.

Codecs introduce heavy non linearity but even then the frequenc
components are moreover preserved. But what i have observed in a rea
recording is that the acoustic echo was present only in the region o
1700-2500 Hz. Now,
1)frequency response of the speaker/mic is not that bad for sure.
2)room reverberations doesn't cause this phenomenon.
3)doesn't make any sense to suspect codecs for it.

The only thing I can suspect is that this acoustic feedback might be du
to the mechanical/acoustic coupling. What do you think ?? and if you hav
an idea, do you think it can be modeled??.

Now that i have given the diagram (hope the situation is clearer now) ca
you suggest any good material which can help me design a AEC for thi
condition.

and Thanks a ton to you guys for responding.

Sankalp

createdon2003 wrote:
> may be it will be more clear by a diagram.
>
> --far_end-------------------|ENC|-------|DEC|--------//
> | | speaker
> | \\
> AEC
> | LEM with
> T_60~60ms
> |
> --to far end----------------|DEC|-------|ENC|--------O microphone
> |

> Now that i have given the diagram (hope the situation is clearer now) can
> you suggest any good material which can help me design a AEC for this
> condition.

You can not. The AEC is not going to work since you have vocoders in the
loop.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

>
>
>createdon2003 wrote:
>> may be it will be more clear by a diagram.
>>
>> --far_end-------------------|ENC|-------|DEC|--------//
>> | | speaker
>> | \\
>> AEC
>> | LEM with
>> T_60~60ms
>> |
>> --to far end----------------|DEC|-------|ENC|--------O microphone
>> |
>
>> Now that i have given the diagram (hope the situation is clearer now
can
>> you suggest any good material which can help me design a AEC for this
>> condition.
>
>You can not. The AEC is not going to work since you have vocoders in th

>loop.

Vlad keeps saying this, and you keep ignoring him, so let add another voi
to this......

Vlad is right. Echo cancellation is a process of system identification
and subtracting the identified system model from the received signal. A lo
bit rate speech codec, like AMR, is heavily lossy. It does not preserve th
signal nearly well enough to allow either the necessary syste
identification or the subtraction process to work.

Steve

steveu wrote:


> Echo cancellation is a process of system identification
> and subtracting the identified system model from the received signal. A low
> bit rate speech codec, like AMR, is heavily lossy. It does not preserve the
> signal nearly well enough to allow either the necessary system
> identification or the subtraction process to work.

Perhaps what the OP really wants is the howling suppression, which is
the different problem then the AEC. The anti howling filter could be
implemented in his setup.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

Yeah I know it very well that I can't model the acoustic path in this case
But for this case also there are solutions (There are products which cater
this problem). It might not be Acoustic Echo Canceler but Acoustic Ech
Control for Non linear channels involving vocoders. There are many pubs o
the same. Neways this was not in my mind to ask when I first made th
posting. My doubt was entirely diff...just drifted away...

But thanks Vlad, Dirk and steve for your time.


Sankalp

On Sep 4, 8:32*am, Vladimir Vassilevsky <nos...@nowhere.com> wrote:
> createdon2003 wrote:
> > may be it will be more clear by a diagram.
>
> > --far_end-------------------|ENC|-------|DEC|--------//
> > * *| * * * * * * * * * * * * * * * * * * * * * * * * | speaker
> > * *| * * * * * * * * * * * * * * * * * * * * * * * * \\
> > * AEC
> > * *| * * * * * * * * * * * * * * * * * * * * * * * * * *LEM with
> > T_60~60ms
> > * *| * * * * * * * * * * * * * * * * * * * * * * * * *
> > --to far end----------------|DEC|-------|ENC|--------O microphone
> > * * * * * * * * * * * * * * * * * ** * * * * * * * *|
> > Now that i have given the diagram (hope the situation is clearer now) can
> > you suggest any good material which can help me design a AEC for this
> > condition.
>
> You can not. The AEC is not going to work since you have vocoders in the
> loop.
>
> Vladimir Vassilevsky
> DSP and Mixed Signal Design Consultanthttp://www.abvolt.com

If that blanket statement were true, then echo cancellers with G.729
vocoders would not work. Nor would channels with ADPCM enc/dec. I
would think that you are obviously restricting your statements to the
high-compression codecs/vocoders. The echo canceller would still
work, just not very well.

Sankalp, echo from the cell side is, indeed, mostly due to the
mechanical feedback from speaker to microphone. There is an ITU
recommendation (can't remember which off hand) that limits this.
However, we have found that not all handsets adhere to this standard.
Worst, the measurement by the manufacturers is highly dependent on how
the test set up was made, which source was used, the artificial head
used, etc. This "echo" is highly non-linear, and linear models do not
represent the echo nor solve the problem well.

You might want to drop a line to Has Gerhlich at Head Acoustics, in
Germany. He has done a lot of research in this area, and in my
opinion, is one of the World experts.

I have often thought that this problem could be eased if one could
somehow efficiently get an adaptive estimate of the Volterra kernel
from the input/output data.

Good Luck, you have certainly pick a very difficult area of research.

Maurice Givens

>On Sep 4, 8:32=A0am, Vladimir Vassilevsky <nos...@nowhere.com> wrote:
>> createdon2003 wrote:
>> > may be it will be more clear by a diagram.
>>
>> > --far_end-------------------|ENC|-------|DEC|--------//
>> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A
=
>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | speaker
>> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A
=
>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 \\
>> > =A0 AEC
>> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A
=
>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0LEM with
>> > T_60~60ms
>> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A
=
>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
>> > --to far end----------------|DEC|-------|ENC|--------O microphone
>> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A
=A0=
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
>> > Now that i have given the diagram (hope the situation is clearer now
c=
>an
>> > you suggest any good material which can help me design a AEC fo
this
>> > condition.
>>
>> You can not. The AEC is not going to work since you have vocoders i
the
>> loop.
>>
>> Vladimir Vassilevsky
>> DSP and Mixed Signal Design Consultanthttp://www.abvolt.com
>
>If that blanket statement were true, then echo cancellers with G.729
>vocoders would not work. Nor would channels with ADPCM enc/dec. I
>would think that you are obviously restricting your statements to the
>high-compression codecs/vocoders. The echo canceller would still
>work, just not very well.

That's an odd thing for someone like you to say.

The non linearities in G.711 limit an echo canceller to about 30 dB or s
of ERLE, with some nasty quirks at low levels that demand NLP.

ADPCM (e.g. 32kbps G.726) reduces the potential ERLE considerably, and NL
has to be used quite aggressively. EC with ADPCM is not all that useful.

You're lucky if you can even make an EC converge through an G.729 channel
and the results are not useful. The quirks are just too severe. You shoul
know that as well as anyone.

Steve

On Sep 8, 11:19*am, "steveu" <ste...@coppice.org> wrote:
> >On Sep 4, 8:32=A0am, Vladimir Vassilevsky <nos...@nowhere.com> wrote:
> >> createdon2003 wrote:
> >> > may be it will be more clear by a diagram.
>
> >> > --far_end-------------------|ENC|-------|DEC|--------//
> >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> =
> >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | speaker
> >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> =
> >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 \\
> >> > =A0 AEC
> >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> =
> >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0LEM with
> >> > T_60~60ms
> >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> =
> >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> >> > --to far end----------------|DEC|-------|ENC|--------O microphone
> >> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> =A0=
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
> >> > Now that i have given the diagram (hope the situation is clearer now)
> c=
> >an
> >> > you suggest any good material which can help me design a AEC for
> this
> >> > condition.
>
> >> You can not. The AEC is not going to work since you have vocoders in
> the
> >> loop.
>
> >> Vladimir Vassilevsky
> >> DSP and Mixed Signal Design Consultanthttp://www.abvolt.com
>
> >If that blanket statement were true, then echo cancellers with G.729
> >vocoders would not work. *Nor would channels with ADPCM enc/dec. *I
> >would think that you are obviously restricting your statements to the
> >high-compression codecs/vocoders. *The echo canceller would still
> >work, just not very well.
>
> That's an odd thing for someone like you to say.
>
> The non linearities in G.711 limit an echo canceller to about 30 dB or so
> of ERLE, with some nasty quirks at low levels that demand NLP.
>
> ADPCM (e.g. 32kbps G.726) reduces the potential ERLE considerably, and NLP
> has to be used quite aggressively. EC with ADPCM is not all that useful.
>
> You're lucky if you can even make an EC converge through an G.729 channel,
> and the results are not useful. The quirks are just too severe. You should
> know that as *well as anyone.
>
> Steve- Hide quoted text -
>
> - Show quoted text -

The original statement was that they don't work. They work, but not
very well. Because they don't converge very well to the desired, is
not to say they don't converge. I'm being very theoretical because I
have a "thing" about blanket statements. There is a lot of
misconception about convergence and stability. You can have the
algorithm converge and not be stable. You can have it be stable and
not converge very well to the desired. You can actually get a LMS-
based algorithm to converge, with small error, to an unstable system.
Analysis of LMS filters typically looks at the weight difference
between the desired and the estimated. If that difference is
relatively large, it is said the filter doesn't converge. It may very
well converge, just not to want you want. LMS is derived from minimum
squared error, not weight difference. This subtlety is often lost
when talking about LMS-based algorithms. When a LMS-based algorithm
produces a relativly stable estimate, it is working, and working the
way designed - minimizing the squared error.

One analysis, using weight difference, looked at adaptive filters in
tandem and came to a conclusion that was arguably not supported by the
data. I did an analysis of the same configuration, but using minimum
squared error analysis to show the result consistent with the data.

Yes, the NLP becomes very "useful" under these conditions. u-law will
converge to about 30 dB ERLE, and a-law to about 34dB. I have had
echo cancellers converge on ADPCM. Not to the 34 dB of a-law, but
they do converge. The reason for mentioning Volterra kernel is that
this is a non-linear model. I have always been interested in how well
a Volterra-based echo canceller would work on these signals (and not
just a 2nd-order Volterra), just never had the time.

Finally, I admit I was nit-picking.

Maurice Givens

On Sep 8, 4:22*pm, maury...@core.com wrote:
> On Sep 8, 11:19*am, "steveu" <ste...@coppice.org> wrote:
>
>
>
>
>
> > >On Sep 4, 8:32=A0am, Vladimir Vassilevsky <nos...@nowhere.com> wrote:
> > >> createdon2003 wrote:
> > >> > may be it will be more clear by a diagram.
>
> > >> > --far_end-------------------|ENC|-------|DEC|--------//
> > >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=A0 =A0 =A0 =A0 =A0 =A0
> > =
> > >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | speaker
> > >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=A0 =A0 =A0 =A0 =A0 =A0
> > =
> > >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 \\
> > >> > =A0 AEC
> > >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=A0 =A0 =A0 =A0 =A0 =A0
> > =
> > >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0LEM with
> > >> > T_60~60ms
> > >> > =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=A0 =A0 =A0 =A0 =A0 =A0
> > =
> > >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> > >> > --to far end----------------|DEC|-------|ENC|--------O microphone
> > >> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
> > =A0=
> > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
> > >> > Now that i have given the diagram (hope the situation is clearer now)
> > c=
> > >an
> > >> > you suggest any good material which can help me design a AEC for
> > this
> > >> > condition.
>
> > >> You can not. The AEC is not going to work since you have vocoders in
> > the
> > >> loop.
>
> > >> Vladimir Vassilevsky
> > >> DSP and Mixed Signal Design Consultanthttp://www.abvolt.com
>
> > >If that blanket statement were true, then echo cancellers with G.729
> > >vocoders would not work. *Nor would channels with ADPCM enc/dec. *I
> > >would think that you are obviously restricting your statements to the
> > >high-compression codecs/vocoders. *The echo canceller would still
> > >work, just not very well.
>
> > That's an odd thing for someone like you to say.
>
> > The non linearities in G.711 limit an echo canceller to about 30 dB or so
> > of ERLE, with some nasty quirks at low levels that demand NLP.
>
> > ADPCM (e.g. 32kbps G.726) reduces the potential ERLE considerably, and NLP
> > has to be used quite aggressively. EC with ADPCM is not all that useful..
>
> > You're lucky if you can even make an EC converge through an G.729 channel,
> > and the results are not useful. The quirks are just too severe. You should
> > know that as *well as anyone.
>
> > Steve- Hide quoted text -
>
> > - Show quoted text -
>
> The original statement was that they don't work. *They work, but not
> very well. *Because they don't converge very well to the desired, is
> not to say they don't converge. I'm being very theoretical because I
> have a "thing" about blanket statements. *There is a lot of
> misconception about convergence and stability. *You can have the
> algorithm converge and not be stable. *You can have it be stable and
> not converge very well to the desired. *You can actually get a LMS-
> based algorithm to converge, with small error, to an unstable system.
> Analysis of LMS filters typically looks at the weight difference
> between the desired and the estimated. *If that difference is
> relatively large, it is said the filter doesn't converge. *It may very
> well converge, just not to want you want. *LMS is derived from minimum
> squared error, not weight difference. *This subtlety is often lost
> when talking about LMS-based algorithms. *When a LMS-based algorithm
> produces a relativly stable estimate, it is working, and working the
> way designed - minimizing the squared error.
>
> One analysis, using weight difference, looked at adaptive filters in
> tandem and came to a conclusion that was arguably not supported by the
> data. *I did an analysis of the same configuration, but using minimum
> squared error analysis to show the result consistent with the data.
>
> Yes, the NLP becomes very "useful" under these conditions. u-law will
> converge to about 30 dB ERLE, and a-law to about 34dB. *I have had
> echo cancellers converge on ADPCM. *Not to the 34 dB of a-law, but
> they do converge. *The reason for mentioning Volterra kernel is that
> this is a non-linear model. *I have always been interested in how well
> a Volterra-based echo canceller would work on these signals (and not
> just a 2nd-order Volterra), just never had the time.
>
> Finally, I admit I was nit-picking.
>
> Maurice Givens- Hide quoted text -
>
> - Show quoted text -


Oops, G.728 is the code-excited linear prediction. This is what I
meant originally, the 16-bit coder based on code-excited LP, not G.
729.

Maurice

[email protected] wrote:


>>The reason for mentioning Volterra kernel is that
>>this is a non-linear model. I have always been interested in how well
>>a Volterra-based echo canceller would work on these signals (and not
>>just a 2nd-order Volterra), just never had the time.

Besides the numerous difficulties such as getting stuck in the local
minima, it would not work because you are dealing with the loss of
information rather then with the dynamic nonlinearity.

>>Finally, I admit I was nit-picking.
> Oops, G.728 is the code-excited linear prediction. This is what I
> meant originally, the 16-bit coder based on code-excited LP, not G.
> 729.

There you go ))) G.728 is a high bitrate engine which behaves similar
to the ADPCM.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com