How is Signal to Noise (SNR) ratio calculated on the receiver?
The SNR of a station on a Loran receiver is calculated as the Loran pulse amplitude, divided by the RMS noise ahead of the pulse (in a region very roughly 50us long) that is observed after pulse subtraction, interferer notching, and ensemble averaging, corrected back to the ADC input. The amplitude is the nominal pulse amplitude, so conceptually it’s measured at the 30uS point, and it’s roughly half the peak amplitude of the pulse. If you ensemble average N Loran pulses, you multiply the amplitude by N but the noise by only sqrt(N), so you improve SNR by N/sqrt(N), which is sqrt(N). This is sometimes called process gain, and it works about the same way in spread-spectrum chipping systems. So to correct the SNR you divide your number by sqrt(N) to take out the process gain again, which refers it effectively to the input. Since the SNR is also inversely proportional to the square root of the receiver bandwidth, and there is no accepted way to normalize for this, SNR will vary somewhat ac
