On Sat, Jul 09, 2005 at 01:33:26PM -0700, Sachi wrote: > > Do you have any suggestion that we can improve the > realibility of the FSK transmisstion? such as > increase the length of the synchronization bits? or > decrease the hamming distance?
The sync pattern was pulled out of a dark place. It should ideally be an m-sequence, or something else with better autocorrelation properties than the random value we choose. See page 15 of this pdf: http://www.ee.columbia.edu/~ywang/MSS/HW2/ss_intro.pdf In the table on page 16, the entry for L=5, gives a 31-bit sequence. We could use one of those sequences padded with a zero to 32-bits. Googling for "m-sequence matlab" will turn up code to generate the sequences. [The sync pattern should be the 31-bit sequence plus a zero]. When you figure out the "improved sync pattern", please let us know and I'll fix the code. Also, note that in the fsk example code there is no error correction or detection. Adding some would be a good idea. Eric _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
