[opendtv] New developments in FEC
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: "OpenDTV (E-mail)" <opendtv@xxxxxxxxxxxxx>
- Date: Wed, 14 Apr 2004 12:51:49 -0400
This article rang a bell. This is a very intriguing concept,
described back in December of 2002 in RFC 3453, where a
new FEC technique is supposed to allow for error-free
file transfer using only FEC, one pass, impervious to
dropouts during transmission, and low overhead. Needless
to say, it would apply to DTV as well as it can to
digital radio.
These codes are generally referred to as "expandable FEC
codes." Here's the description of how it works, from RFC
3453, Section 2.5:
"From any k of the received encoding symbols, the FEC
decoder recreates the k original source symbols as follows.
If all k original source symbols are received, then no
decoding is necessary. Otherwise, at least one redundant
symbol is received, from which the receiver can easily
determine if the block is composed of variable- length
source symbols: if the redundant symbol(s) is longer than
the source symbols then the source symbols are variable-
length. Note that in a variable-length block the redundant
symbols are always longer than the longest source symbol,
due to the presence of the prepended symbol- length. The
receiver can determine the value of lmax by subtracting x
from the length of a received redundant symbol. For each
of the received original source symbols, the receiver can
generate the corresponding padded source symbol as
described above. Then, the input to the FEC decoder is
the set of received redundant symbols, together with the
set of padded source symbols constructed from the received
original symbols. The FEC decoder then produces the set
of k padded source symbols. Once the k padded source
symbols have been recovered, the length l_i of original
source symbol i can be recovered from the first x bytes of
the ith padded source symbol, and then original source
symbol i is obtained by taking the next l_i bytes
following the x bytes of the length field."
Claro?
Bert
------------------------------------
Deal pushes algorithms into digital radio
Loring Wirbel
Apr 13, 2004 (11:00 AM)
URL: http://www.commsdesign.com/showArticle.jhtml?articleID=3D18901216
COLORADO SPRINGS, Colo.- Digital Fountain Inc. will announce
licensing pacts this week with Honda Motor Co. and XM Satellite
Radio Inc. that will push its data-transport algorithms into
digital radio.=20
The Raptor algorithms, based on the company's patented Luby
Transform Codes, promise full data transmission without packet
retransmission, essentially replacing TCP transport with what
the company calls "metacontent" packets.
Honda and XM have not said how Raptor will be used, but Digital
Fountain president and chief executive officer Charlie
Oppenheimer said that auto manufacturers want to use satellite
radio as a baseline service for providing "Tivo-like" functions
within the in-car network, including video storage and
retransmission, global-positioning satellite coordinate updates
and Internet data services. Synchronizing such services would be
next to impossible in normal terrestrial cellular or satellite
networks, since moving vehicles go through dead zones.
Oppenheimer said that the work of the company's founder,
University of California at Berkeley professor Michael Luby, is
similar to math principles on the solution of simultaneous
linear equations, implying that receipt of packets in an
aggregate length equal to the size of the original transmission
would allow reconstruction of the message, regardless of whether
packets were dropped.
"This eliminates acknowledgement and re-transmission, because it
doesn't matter which packets you receive," he said.
The algorithm can be embedded into traditional codecs, but is not
intended as a replacement for compression. In fact, Sumitomo
Electric Co. has licensed the Digital Fountain technology for use
with MPEG-4 codecs for its StreamCruiser Internet Protocol-based
set-top box. The metacontent packets essentially eliminate the
need for forward-error-correction circuits, which could radically
change the design of broadband wide-area network line cards if
the algorithms are widely adopted.
The metacontent packet creator, which can be implemented in a
server or other centralized hardware system, uses UDP unicast or
multicast to send the packets to other nodes in the system.
Oppenheimer said that Digital Fountain anticipates using its
algorithms as a basic Layer 4 transport technology for a variety
of future markets.
"Another area we're seeing immediate interest in is transmission
of video over 802.11 wireless networks," Oppenheimer said.
Transmission of loss-free video can be implemented over Wi-Fi,
digital subscriber line or cable, he said, and over any virtual
private network operating at Layers 2 or 3.
Copyright =A9 2003 CMP Media
----------------------------------------------------------------------
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:
- » [opendtv] New developments in FEC
