SNR Estimation for a Non-Coherent M-FSK Receiver in a Slow Flat Fading Environment

Estimation of the signal-to-noise ratio (SNR) is considered for a non-coherent M-ary frequency shift keying (NC-MFSK) receiver. It has been assumed that the transmitted symbols undergo a slow flat fading channel where a block of data is corrupted by an independent constant fade and additive white Gaussian noise. Two approaches for SNR estimation are reported in this paper: an approximate maximum likelihood approach and another using the data statistics, both for data aided and non-data aided systems. It has been shown that for a particular SNR region of interest and depending upon the availability of pilot symbols, a particular approach is suitable for SNR estimation.