Hello everyone, I have a problem with OFDM frame detection for one of my channels. I work with gr-ieee802-11 blocks and receive from 4 channels using X300 and 2 TwinRX. For the frame detection, the receive blocks calculate the autocorrelation of the incoming stream over a specified window size and divide it by the average power over this window. After that, the normalized autocorrelation value is compared with a threshold to determine whether there is a frame or not and 'wifi_start' tag is inserted if there is a frame.
This is what I see the FFT output of the channels( https://imgur.com/a/UeXTU ). For the other channels, I see this wifi_start tag with a period corresponding to my interval set by the transmitter. But the channel 1 which is A:1 of my USRP shows the tag all the time. When I checked the correlation value of the channels, I saw that this channel always has a correlation value around 0.7( https://imgur.com/a/KHgXn ). It shows wifi_start tag since I have threshold 0.56. When I increase the receiver gain of this channel, the normalized correlation value is decreasing but the constellation gets more scattered with high receive gain and it s hard to decode with this constellation. I set my threshold 0.75 for this channel, I was able to decode but it did not work as properly as the other channels(missing frames, high frame error etc.). Do you have any guess about the issue? Why do I have a high correlation value for this channel although the other channels have similar values to each other? Can I assume that there is something wrong with the receive chain of this channel? Best, -Bugra
_______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
