Hi Marcus, i attached the flowgraph to this mail. The constellation Modulator places the tx_eob tag too early (sps*8/ ld(QPSK.arity()) ), which is fixed by using a oot-module, delaying it.
Using BPSK instead of QPSK works fine, so I think the problem is located in the way using the correlation estimator, but i am not sure about it. I am looking forward to hearing from you and thank you in advance. Sebastian Von: Sebastian Gesendet: Montag, 28. August 2017 17:13 An: Marcus Müller Betreff: AW: [Discuss-gnuradio] Using Correlate Estimation with hardware Hi Marcus, i attached the flowgraph to this mail. The constellation Modulator places the tx_eob tag too early (sps*8/ ld(QPSK.arity()) ), which is fixed by using a oot-module, delaying it. Using BPSK instead of QPSK works fine, so I think the problem is located in the way using the correlation estimator, but i am not sure about it. I am looking forward to hearing from you and thank you in advance. Sebastian Von: Marcus Müller Gesendet: Samstag, 26. August 2017 18:34 An: discuss-gnuradio@gnu.org Betreff: Re: [Discuss-gnuradio] Using Correlate Estimation with hardware Hi Sebastian, well, it would sound like you've discovered the multipath channel! I don't really understand what you mean with "Modulate vector is called"; what /is/ "Modulate vector"? You can actually simply attach small files to emails to the list. Such a simple GRC file would certainly not upset us :) Best regards, Marcus On 26.08.2017 18:10, Sebastian wrote: Hi all, i run into trouble using the correlate estimation block. As far as i understand this blocks is used to search for a praeamble which is modulated and filtered with match filter. To achieve this i used modulate vector. I used this block like it`s done in the corr_est exapmles provided in /gnuradio/gr-digital/examples/. Using no UHD but a channel modell everything works fine having clear correlation peak. Using it with hardware device it doesn`t work anymore, having more peaks avaiable. My flowgraph is very simple: Vector_source (using prae+data as vector tagged with tx_sob and tx_eob) -> constellation Modulator(default-QPSK-object) -> stream mux with null_source for break -> uhd sink(sync = PC_clock) Uhd source (sync = PC_clock) -> low_pass -> corr_est(modulated_vector, 0.9) -> Pll-clock-sync -> Costas Loop -> constellation-demodulator -> float_to_char -> qt time sink Modulate vector is called with default-qpsk-object as constellation object and [1] as filter taps which i think is right. The praeamble i am using is the one provided in the examples, so it should be ok too. I dont have that much experience in this topic so I hope somebody is able to help me. I am looking forward to hearing from you and thank you in advance. Sebastian PS : I am not sure where and how to provide the flowgraph itself but i´ll try to do it asap. _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
mytest.grc
Description: Binary data
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
