The output file that I save contains I (Real) and Q (Imaginary) parts of the received signal after passing mixer from which the phase can be easily extracted, phase=tan(Q/I). While the phase can be easily extracted from above, RSSI parameter is measured for phase measurement. Why not directly using the measured I and Q and rather using RSSI parameter?
If you mean streaming out the I and Q, performing the phase extraction on the modem, I don’t see why not. The advantage of the former is that it takes less (data output) bandwidth and supposedly you can average the phase reading with the on chip MCU (but requires programming the thing). I am going for the I-Q stream ultimately because one can perform a better analysis of the statistics. My question is what is the advantage in setting the I channel gain to unity (and using sin) as opposed to tan(Q/I)? They both appear to work.