search

git-artes / gr-isdbt

DTV ISDB-T in GNURadio
http://iie.fing.edu.uy/investigacion/grupos/artes/gr-isdbt/
Other
148 stars 35 forks source link

Weird BER Measurements #42

Open ceeae10 opened 4 years ago

ceeae10 commented 4 years ago

I'm varying the transmitter and receiver gain to show the differences in signal power, constellation clarity, MER measurements, and BER Measurements. However, the BER Measurements I have been getting is at -18dB. I have read that the expected value for BER is at 2x10^-4(around -3.698dB). Is this an error? Because I find having an -18 dB hard to believe.

Thanks for your help

git-artes commented 4 years ago

Hi, If you are getting -18 dB it means that you have zero error (not sure why, probably internally GNU Radio adds a little epsilon to the log). Is this is a simulation? Is it the Viterbi or Reed-Solomon BER? best

ceeae10 commented 4 years ago

This is wireless transmission. It's the Viterbi BER. The change was drastic from around ~-3dB at lower signal power and then when increased, it shoots up to -18dB. From there, it randomly fluctuates. See the picture for reference.

This is RX Gain at 30 dB and TX Gain at 65 dB Reference

git-artes commented 4 years ago

Then you are getting zero error during your high-power transmission. The recording we share at the project's website has a BER of 10^(-9) or less, as we had line-of-sight with station's antenna. In this case, where I am assume you are transmitting next to the receiver, it's not rare that you get a virtually flawless signal.