[opendtv] Re: Ofcom paves the way for UK HD on DTT
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Sat, 5 Apr 2008 06:52:50 -0400
At 3:48 PM -0400 4/4/08, Manfredi, Albert E wrote:
Craig Birkmaier wrote:
> But is the Ofcom proposal using MIMO?
DVB-T2, so yes.
Are you sure? MIMO is a very different approach that adds complexity
to both the transmission infrastructure and to the receivers. I do
not think it is required for DVB-T2, and it was not mentioned in the
article you linked to.
We don't know just how easy it is with H.264, nor what kind of HD they
are planning to psovide in those 4 HD channels. All I do know is that
WETA-DT transmits 3 SD streams and one HD stream in a 6 MHz multiplex,
and unless the HD stream is trying to do rapid motion, it works quite
well.
Actually we do Bert. 12 Mbps is adequate even for fast motion with
h.264. 12 Mbps is barely adequate for low motion 24P with MPEG-2 and
totally inadequate for ANY 30i/60P source ( not that they don't try).
The WETA multiplex aqssures that all streams are bit starved and will
show significant artifacts, OR require significant pre-filtering to
reduce encoder stress.
Again, think a bit more like an engineer, less about hype, for a change.
Slow down and think logically.
DVB-T2 depends on keeping an H and a V path separated, to attempt to
provide twice the bandwidth in the same spectrum. Their tests have shown
that this is not always possible, in different clusters of the coverage
area, and that to compensate for the problem areas, they have to
increase the C/N margin in their planning factors.
This is ONLY for MIMO. The rest is not much different that DVB-T
other than the increase in the number of carriers.
MPH goes exactly opposite of this. It looks for maximum robustness, i.e.
lowest possible C/N. Instead of ~24 db C/N required for trying to
achieve reliable DVB-T2, and still with pockets in the area where double
the bit rate isn't possible, MPH is looking at 1.9 to 4 dB of C/N.
I see that the spectral efficiency is reduced to make robust
operation possible with VSB. We already demonstrated equivalent
mobile reception in Las Vegas, in 2000, with plan vanilla DVB-T, with
a payload that was equivalent to 8-VSB without any MPH enhancements.
As MIMO required dual antennas, it would be interesting to see if a
proper receiver can be designed to work with it in mobile
applications.
Regards
Craig
----------------------------------------------------------------------
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:
