SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : C-Cube
CUBE 36.04-0.8%Dec 31 3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: DiViT who wrote (21314)8/25/1997 2:20:00 PM
From: BillyG   of 50808
 
CUBE listens to its customers..........

Another excerpt from the white paper. CUBE helps reduce costs for customers that rent satellite transponders at about $2 million per transponder per year.

Bandwidth Utilization Techniques
Bandwidth is expensive. On average, it costs about $2 million a year to operate
a satellite transponder. Terrestrial broadcasters have been allocated just one 6
MHz channel for transmitting all their video programming. Before digital
compression technology, one satellite transponder or one 6 MHz terrestrial
channel could carry only a single video program. Compression enables that
same transponder or 6 MHz spectrum to carry multiple channels-typically 4 or
5 fixed bit rate channels.

It has recently been shown that the number of channels within a given bandwidth
can be increased without loss of video quality by implementing a key feature
called Statistical Multiplexing, or stat-mux. Multiplexing is the method whereby
data from multiple channels is combined into a single bitstream and transmitted
together. The term "statistical" refers to the fact that, statisically, various
channels operating simultaneously will not all contain difficult-to-encode material
at the same time.

One goal of the basic MPEG-2 compression algorithm is to maintain a constant
bit rate for a particular channel within the multiplex. Therefore, if a transponder
could transmit 30 Mbps, then a service provider could transmit five channels,
each operating at 6 Mbps. Or the operator might opt for three channels
operating at 4 Mbps and two channels operating at 9 Mbps.

In any case, the data rate allocated to each channel remains constant, which
could cause the quality of the decoded video streams to fluctuate slightly. The
quality might improve if easy-to-encode material was presented, or it might
degrade slightly when the encoder is presented with difficult material. Although
the encoder can modify the bit rate it generates slightly in order to maintain
quality, it must always keep its output buffer filled, but not overflowing. The
encoder is therefore constrained by the rate at which the output channel drains
its buffer.

The purpose of statistical multiplexing is to keep the quality constant while
allowing the bit rate for each channel to fluctuate-i.e., the rate at which each
encoder's buffer is drained will fluctuate. The more channels that are multiplexed
together, the more likely it becomes that some channels will display
easy-to-encode material using fewer bits, while other channels display
hard-to-encode material requiring more bits. Overall, the sum of the bit rates
should equal the transponder rate. Figure 4 gives a graphical depiction of
statistical multiplexing.

In order to accommodate this type of algorithm, encoder systems must be able
to measure the quality of the encoded video they are producing, and report
these results to a system controller. In addition, encoders must be able to
change the output bit rate they generate very quickly in response to commands
from the system controller. The architecture described in this paper can vary the
bit rate within 1/60th of a second, and is capable of providing quality estimates
on a frame-by-frame basis to the system controller.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext