Single channel source separation and parameter estimation of multi-component PRBCPM-SFM signal based on generalized period

Abstract

This paper proposes an approach of single channel source separation and parameter estimation of multi-component PRBCPM-SFM (pseudo-random binary code phase modulated sinusoidal frequency modulation) signal. By transforming all components to SFM signals through square calculation, we may then apply singular value decomposition to determine the generalized period and the modulation frequency of the SFM signal. The modulation index and carrier frequency are determined by searching for discrete points and optimizing calculations. The initial phase can be determined by calculating the inner product. Finally, given that the pseudo-random binary code is a real signal, the PN (Pseudo-noise) sequence and the amplitude can be estimated. Our proposed method can estimate the SNR by using the method of subspace-based decomposition, and the estimated SNR can be used to adaptively stop separation. Experimental results demonstrate the algorithm's performance.