On Thu, May 10, 2012 at 3:12 AM, Andreas Weller <[email protected]> wrote: > On Wed, May 9, 2012 at 7:28 PM, Bruce Perens <[email protected]> wrote: >> Hm. Or stay with the lower bit-rate implementation and add LDPC? > > I agree on that. This way the interaction of codec, FEC and modem > could be optimized. > > May be one of these LFPC GPL projects could be used to prevent David > from reinventing wheel: > http://planete-bcast.inrialpes.fr/article.php3?id_article=16 > http://www.cs.utoronto.ca/~radford/ldpc.software.html
The planete-bcast stuff is erasure codes, not really want you want here. The software at the second link is useful. A fun thing to do that wouldn't require too much thought would be to generate random LDPC matrixes (constrained to ones that will actually work), and then test them with the actual modem, a path simulator, and the codec2 decoder and measure the resulting quality with something like PESQ (hmm, does PESQ work on codec2?). Then just brute force search for the check matrix that gives the best results. ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
