Sparsity maximization nonlinear blind deconvolution and its application in identification of satellite microvibration sources

In this study, sparsity maximization nonlinear blind deconvolution (NBD) is proposed to identify the vibration sources of satellite systems from mixed vibration signals. The proposed algorithm decomposes NBD into two independent stages, namely, nonlinear compensation and blind deconvolution. Since nonlinear distortion weakens the sparsity of the observed signals, sparsity maximization is introduced to the nonlinear compensation stage. In the blind deconvolution stage, the blind deconvolution algorithm with reference is used to separate the source signals. The proposed algorithm can improve the accuracy of source signal extraction from nonlinear mixed signals of complex mechanical systems. The effectiveness of the proposed method is verified through simulations. An experimental system of aluminum cabin structure is built based on the satellite’s cabin structure. Results show that the proposed algorithm can successfully realize the identification of source signals.