The Signal-to-Noise Ratio Estimation Techniques for PCM Signals

Abstract

Reliable estimation of signal-to-noise (S/N) ratio in a demodulated PCM telemetry signal can be useful in evaluating the performance of the complete telemetry link, including its signal detection and data processing portions. This paper describes three potentially practical methods developed at Goddard Space Flight Center for estimating the S/N ratio in a PCM signal. One method referred to as "spectral null" method uses spectral characteristics of PCM' signals to estimate the S/N ratio, the other two use statistical properties of the signal, i.e., its mean value and variance. These two methods are known as "variance estimations and "null zone." The implementation of each method is discussed. The spectral null method takes the least amount of equipment, but is more difficult to calibrate and operate over a wide range of bit rates, than the other two systems. All three approaches, however, are uncomplicated enough to be included into almost any existing PCM data handling system. An analysis of the performance characteristics of each system is made. It is shown that the variance estimation method is the most versatile. It can reliably estimate the S/N ratio to within 1.5 db over a range of S/N ratios from 0 db to +10 db. (The S/N ratio is defined as the ratio of signal energy per bit/noise power density.) Under certain conditions all three methods can provide estimates to within 1 db, especially over a S/N ratio range from +3 db to +10 db.