As briefly mentioned at FOSDEM, I have been investigating RDS since last christmas, trying to tackle each issue step by step from LO synchronization for BPSK demodulation, bit decoding, sentence synchronization, station name and free text extraction, error correcting code implementation and finally transferred time analysis. My objective was of course not to compete with gr_rds, used as a reference, but to acquire a detailed understanding of each step. Since I protoype with GNU/Octave, I wanted to convert most digital signal processing from LFSR to linear algebra (matrix). I think I pretty much understand now the whole process, despite designing the error correcting code being still beyond my capability. An internal lab presentation on the topic is now available at http://jmfriedt.free.fr/slides_rds.pdf and for the French speaking audience, accompanying article (in French at the moment, will be translated ASAP to English, as soon as I have completed RDS v.s GPS time comparisons) at http://jmfriedt.free.fr/lm_rds.pdf
I am well aware of Tom Rondeau's sponsoring Polyphase clock recovery, but at the moment I could only figure out how to use MM-Clock recovery and MPSK, which worked quite well for my purpose. Any feedback welcome, JM -- JM Friedt, FEMTO-ST Time & Frequency/SENSeOR, 26 rue de l'Epitaphe, 25000 Besancon, France _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
