[opendtv] Re: Integrated DTV PSIP
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: <opendtv@xxxxxxxxxxxxx>
- Date: Sun, 4 Dec 2005 18:33:08 -0500
> > 2K COFDM, at 64-QAM, should be roughly equivalent to a
> > Linx receiver.
Bob Miller wrote:
> Not even close. Have you ever tested a Linx receiver?
Presumably, then, you have a well documented set of
comparative test results, which I'm always happy to
review.
Failing that, in terms of static echo tolerance, here are
the numbers.
2K COFDM in a 6 MHz channel has a symbol duration of 299
usec. At equal spectral efficiency, you'd be using 64-QAM
and a guard interval of 1/16. Because in a 6 MHz channel,
with 3/4 FEC, this results in 19.76 Mb/s.
Therefore, the guard interval consists of 299/16 or 18.7
usec of time during which echoes are permitted to appear,
with respect to the main signal.
You'll find this compares quite favorably with the Linx
receiver even at 0 dB echo levels, but the Linx can exceed
these values at lower echo strength. The Linx, at 0 dB
echo strength, can tolerate echoes 17 usec before or after
the main signal.
Dynamic echo is another matter, though, because COFDM
should not be as handicapped as a system that depends on
an equalizer to track the changing echo. Also, the Linx
didn't use the training sequence of 8-VSB, relying instead
on blind equalization only, which probably conspired to
make it less than stellar in terms of dynamic echo.
Also another factor would be the basic effects of tuner
selectivity and dynamic range. But those considerations
apply equally to COFDM or 8-VSB receivers. I don't know
if the Linx prototypes used cheap tuners or not.
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:
