RE: [ecc] The soft decision decoding and else ...

["Li Kwet Pin Chi Van" <ECVLi@ntu.edu.sg>]

Hi all,

That's great to see that I am getting mail from this list from different
ppl:) 

Zalfany: I have asked Dali if he could change your Viterbi core from
hard to soft decision. Let me give you some of the main features of the
Viterbi core according to IEEE802.11a. It decodes a K=7, G=[171, 133]
convolutional code. However it is the speed that is of concern since the
max data rate of this WLAN is 54Mbps. Right now, I have a Virtex in my
possession but hope to get a Virtex II soon (apparently it looks like
there is a shortage on the market at the moment). WIth such a part,
maybe we can easily go to a 300MHz clock. At this speed, we will have
about 6 clock cycles to perform the decoding.

Now I am working on a RS codec. Anybody wants to share or to ask
something about this codec, you are most welcome:)

Best Regards,
Chi Van


-----Original Message-----
From: Zalfany Urfianto [mailto:zalfany@yahoo.com]
Sent: Saturday, November 10, 2001 6:27 AM
To: ecc@opencores.org
Subject: Re: [ecc] The soft decision decoding and else ...


On 9 Nov 2001, at 21:34, mohamed.kilani_ecc@m4x.org wrote:

> BTW, I will try to write a scope explaining the Soft Decoding Approach

> so we can work together on creating an architecture for this decoding 
> method 

OK. Your scope explanation might be a good start for discussions on 
this maillist. Some body has to start heating up people freezed 
somewhere inside this list :-)
 
> If you can go ahead and pipeline your design and try to synthesize it
(if 
> you have the tools to do that, IcarusVerilog can do that ..). 

There must be some way for me to get the access to synthesis tools 
and implement them at campus. I'll try to get one so that I have 
something to do when all my friends go home for christmas.

> There is a  WLAN design project that needs a throughput of 54Mbps so
we have a 
> lot of work to do !!!!

Wow !! Could you describe more on what kind application is it going 
to be implemented ?

> I already changed your code to handle the used polynomial for this 
> standard. The design compiles but some bugs are still there and I
didn't 
> have time to fix them yet, perhaps this WE ...

Which part. Who knows I might help.

> You are saying that you don't know what to do now ...
> Don't worry if you wanna be busy, I'll keep you so :))

OK. It does not mean that I have lot of spare time, but I'll try to 
catch up with the assignment you are given then.

> We can start by this soft decision thing and then we can implement APP

> decoding method (better than Viterbi) and then implement a turbo-code 
> encoder/decoder ultimately and even an LDPC encoder/decoder if 
> someone needs any of these ... 

APP decoding, never heard of it :-). Turbo code, AFAIK, is surely 
better than convolutional code. LDPC .. the same as for APP decoding.

> I hope we will have volonteers to help out with all these nice ideas
:-)

Just show that we can create something useful, then people will start 
coming by itself I guess.

By the way, for the last 4 months, I've got inquiry from someone in 
India who is trying to implement the Viterbi decoder too. Not sure 
that his project could be shared among the Opencores community, but 
he still have some troubles too, so maybe we just have to wait a 
little longer.

Regards,
- Zalfany

_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

--
To unsubscribe from ecc mailing list please visit
http://www.opencores.org/mailinglists.shtml
--
To unsubscribe from ecc mailing list please visit http://www.opencores.org/mailinglists.shtml