Yes, for sure. That's the property of short training sequence I think. In SORA and my code we have already saved frequency domain values of the short training and long training sequence with which we correlate the incoming samples and look for sharp peaks and then count the peaks.
For my code it comes something like this <http://gnuradio.4.n7.nabble.com/file/n61940/Screenshot_from_2016-11-08_20-37-24.png> The multiplication and addition stuff we do with SSE instructions, so its pretty fast, but have to compare it with SC based methods ! Sumit -- View this message in context: http://gnuradio.4.n7.nabble.com/ofdm-sync-sc-cfb-impl-cc-and-sync-methods-used-in-practical-tp61933p61940.html Sent from the GnuRadio mailing list archive at Nabble.com. _______________________________________________ Discuss-gnuradio mailing list [email protected] https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
