[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

HDTV on the "small" screen

The economics surrounding the design of receivers capable of multiple
transport stream demodulation and multi-channel decoding present additional
arguments favoring COFDM and layered encoding over 8VSB.

With the newly available high-speed SiGe DSPs and COFDM, a single
demodulator chip could perform the work of multiple tuners.  The unique
aspect of CODFM is its use of thousands of separate carrier frequencies for
each 6 MHz transport stream channel.  Instead of sampling a single carrier
frequency several million times per second as with 8-VSB, a COFDM
demodulator samples two thousand to eight thousand carriers a few thousand
times per second.  The DSP chip simply tunes to a different carrier between
samples when demodulating a broadcast signal.  With COFDM's far greater
latitude than 8-VSB for sampling the carrier signal, a single chip can
simultaneously sample and demodulate other carriers in other 6 MHz frequency
bands.  Multiple transport stream demodulation for COFDM therefore would
likely require only a few modest changes to the demodulator chip software
and hardware design.  Although stream buffer memory would need to be
expanded, the extra cells could still be easily included on a single chip.

The economics favoring layered encoding with multi-channel decoding is
equally compelling.  An HD1-level chip capable of decoding a single 720p
video channel has sufficient processing power and memory buffer space to
decode three separate 480p channels.  If the 720p image is derived from two
streams under a layered encoding approach, specifically a 480p base and a
720p enhancement layer, the viewer could opt to watch only the 480p base
layer and up to two other 480p streams displayed in PIP or POP format.
These other 480p streams could be the base layers as well for other 720p
channels.  Obviously, this approach requires a relatively fast scaling
engine, but its work could be simplified and reduced by the choice of
scaling factors, 2:1 or 4:1 for example.  In contrast, multi-channel
decoding without a layered encoding approach would need three separate
HD1-level decoders and even faster scaling engines for the same
functionality.  The only downside with this layered encoding approach to
multi-channel decoding would be a slight loss in the resolution of the
primary picture.  The difference in PIP or POP picture quality would
probably be imperceptible if not better under this cheaper alternative.

thanks to Aine Marsland and Pandora International for support in 1999
No advertising/marketing allowed on the main TIG.  Contact rob at alegria.com
1029 subscribers in 40 countries on Sat Jan 23 18:24:42 CST 1999 
subscribe/unsubscribe with that Subject: to telecine-request at alegria.com
complete information on the TIG website http://www.alegria.com/tig3/
anonymous messaging now at http://www.alegria.com/HyperNews/get/ubique.html