Time-frequency peak filtering applied to FSK signals

This paper considers the problem of non-coherent detection of M-ary frequency-shift-keyed signals embedded in white Gaussian noise. The time-frequency peak filter (TFPF) is suggested as a technique to improve the signal-to-noise ratio of these signals before detection. First the method of TFPF is briefly reviewed and it is shown that TFPF is in general biased when the peak of the Wigner-Ville distribution (WVD) is used to estimate the signal instantaneous frequency. Then criteria are given for determining the reduced bias pseudo WVD window length which allows for practical implementation. Finally we demonstrate using simulations, the effectiveness of TFPF for improving non-coherent detection of MFSK signals