Blind Separation of Synchronous-networking Frequency Hopping Signals based on Time-frequency Analysis

A novel method on smoothed-pseudo Wigner-Ville Distribution (SPWVD) to extract single-source zone of the synchronous-networking frequency hopping (FH) signals is proposed in this paper. Since the hop rate and hop period of synchronous-networking FH signals are the same, the method is applied combining with these characteristics. Compare with other methods such as independent component analysis (ICA), the algorithm takes the advantage of the sparse distribution of FH signal in frequency. Then the blind separation algorithm of FH signals with time-frequency ration of matrix (TIFROM) based on short time Fourier transform (STFT) is improved. Simulation results show that the proposed algorithm is effective in the separation of synchronous-networking FH signals, even though the interval between FH signals is relatively small.