Hey Julius, the OFDM codes work over the air, so I don't think statefulness is the issue. Maybe you just have super-full buffers or something like that.
The packet detection is the most stable part -- can you see if that works? Just use the Schmidl&Cox block from the rx and see if it prints tags? M On 14.07.2015 08:26, Julius Durst wrote: > Hi list, > > I want to reconfigure an OFDM system. For this, I am so far simply > trying to reconnect the tx and rx in the loopback example from the > gr-digital ofdm examples. I made only very small modifications to it > with two selector blocks, so now the flowgraph looks like this: > http://i.imgur.com/Otr0VUt.png > grc file: http://pastebin.com/pTDFhgvq > > As soon as the OFDM rx is not connected from the beginning of execution > or is reconnected after being disconnected, it is not able to receive > the data. > > When being disconnected and reconnected again, one of the following > scenarios occurs: > > 1. Nothing happens, just nothing more is received: > > ---------------------------------------------------------------------- > Tag Debug: Rx Packets > Input Stream: 00 > Offset: 24150 Source: n/a Key: packet_num Value: 483 > Offset: 24150 Source: n/a Key: ofdm_sync_carr_offset Value: 0 > Offset: 24150 Source: n/a Key: rx_len Value: 50 > ---------------------------------------------------------------------- > >>>> Done > > > 2. Sometimes I get output like this: (Is there a list with return codes? > What is -9? Is it normal when killing flowgraph with the button in grc?) > > ---------------------------------------------------------------------- > Tag Debug: Rx Packets > Input Stream: 00 > Offset: 33700 Source: n/a Key: packet_num Value: 674 > Offset: 33700 Source: n/a Key: ofdm_sync_carr_offset Value: 0 > Offset: 33700 Source: n/a Key: rx_len Value: 50 > ---------------------------------------------------------------------- > INFO: Detected an invalid packet at item 32400 > INFO: Parser returned #f > > >>> Done (return code -9) > > > 3. The return code seems to not necessarily having to do with the INFO, > because also got this: > > ---------------------------------------------------------------------- > Tag Debug: Rx Packets > Input Stream: 00 > Offset: 26100 Source: n/a Key: packet_num Value: 522 > Offset: 26100 Source: n/a Key: ofdm_sync_carr_offset Value: 0 > Offset: 26100 Source: n/a Key: rx_len Value: 50 > ---------------------------------------------------------------------- > INFO: Detected an invalid packet at item 25104 > INFO: Parser returned #f > > >>> Done > > > 4. When first connecting the null sink and then connecting rx to the > already running tx: > > Using Volk machine: avx_64_mmx_orc > INFO: Detected an invalid packet at item 0 > INFO: Parser returned #f > INFO: Detected an invalid packet at item 48 > INFO: Parser returned #f > INFO: Detected an invalid packet at item 96 > INFO: Parser returned #f > INFO: Detected an invalid packet at item 144 > INFO: Parser returned #f > INFO: Detected an invalid packet at item 192 > INFO: Parser returned #f > INFO: Detected an invalid packet at item 240 > INFO: Parser returned #f > ... and so on > > > Note that the ">>> Done" is from manually killing the flowgraph, it did > not stop by itself, just added it because of the exit code. > > So my question is: Why is this not working? In the cases where I just > did not get any new packets (1., 2. and 3.), it seems like the Schmidl & > Cox block does not even detect a new packet, so maybe something is wrong > with that block? I also checked the "detected" output of the Schmidl & > Cox block, it only had zeros. > > Thanks in advance for your help and ideas > > Julius > > _______________________________________________ > Discuss-gnuradio mailing list > Discuss-gnuradio@gnu.org > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio