[opendtv] Re: DTT tuner design
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: <opendtv@xxxxxxxxxxxxx>
- Date: Thu, 5 Jul 2007 20:39:49 -0400
Dale Kelly wrote:
> That is, assuming the combined trellis and R-S FEC
> schemes are more cleverly exploited as you suggest.
> The potential is clearly there but, do you believe
> that this will actually happen in the US system? I
> know that I'm a pessimist but it is for good
> reasons.
Dale, first, to be clear, the numbers I showed for the Gemini chip were
the numbers the CRC reported. So there is nothing new in how the two FEC
schemes are used there. Those C/N figures show that echo energy is being
used as part of the usable signal. The fact that the baseline is still
15 dB of C/N shows that trellis and R-S are used in series.
As to your question, if you don't have to change the transmission
standard, then the improvement should be easy to introduce. Each
manufacturer can choose to make the improvement on his own schedule.
In this case, a known improvement today would be to use the trellis code
to pinpoint where the error occurred, indicate that to the R-S code,
then use R-S only to correct the error bytes. That alone will allow
correction of twice as many bytes as what is possible now (20 bytes vs
10 correctable per 208/188-byte frame). Al mentioned this several times,
and it's mentioned in one of the ATSC standards now. I think A/53.
The effect that has is to edge down the C/N margin before errors become
noticeable.
Changing the basic FEC scheme to a turbo code would be much more
difficult. A-VSB cleverly introduces turbo code, by layering it on top
of the existing standard. And it gets very close to the Shannon limit
that way. But I'm not dismissing the possibility that other clever
things can be done with the existing transmission standard. Unleash
enough math PhD candidates on the problem and see what comes out.
I'm only advocating making the best with what we have, for the wide
stream, for the time being.
Bert
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
