[opendtv] Re: 20060912 Twang's Tuesday Tribune (Mark's Monday Memo)
- From: Richard Hollandsworth <holl_ands@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Thu, 14 Sep 2006 21:26:37 -0700 (PDT)
Who makes Diversity Antenna Processor for 8VSB???
Although they are very plentiful for DVB-T, I have yet to run across one for
8VSB, other than in some lab resuts from CRC and Korea researchers. Must be a
much easier signal processing job for COFDM....
holl_ands
==================================================
Albert Manfredi <bert22306@xxxxxxxxxxx> wrote:And you didn't compare this with,
say, an LG 5th gen or Samsung Gemini
system.
Here's what the numbers say SHOULD happen, assuming the same spectral
efficiency. And this is why I'd like to see some real test results from
China.
The COFDM receivers will work under a wider range of multipath scenarios. So
they will appear more robust than even the Samsung Gemini, but not by a
whole lot. (The Gemini goes out to +60 usec, which is about like COFDM at
1/16 GI. But not as good as COFDM under strong pre-echo.)
But when the multipath is within the limits of the 8-VSB demod, the 8-VSB
receiver should be able to decode a weaker signal successfully.
If you're in severe multipath, even Brazil E, a good VSB receiver should
work well. Brazil E is three equal strength paths, 1 usec apart. If you
spread out the Brazil E echoes more and more, eventually the COFDM receiver
will do better.
Similarly, if you have no diversity antennas, COFDM will do better with deep
nulls. With diversity antennas, 8-VSB will catch up some and perhaps even
beat COFDM, again depending on the echo range.
And for plain old long range reception, where multipath is usually too weak
to be an issue, 8-VSB should do better by several dB. Then again, this last
part was even true with 1st gen 8-VSB, as Sinclair found out. Should be
better now.
Bert
---------------------------------
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail.
Other related posts: