A hybrid diagnosis method for defective array elements based on compressive sensing and Iterative shrinkage thresholding algorithm

The aim of this conference paper is to introduce a novel and effective diagnosis method for defective array elements by hybrid Compressive Sensing (CS) and Iterative Shrinkage Thresholding Algorithm (ISTA). The proposed method utilizes the framework of Compressive Sensing and the sparse signal is constructed via difference incentives of reference array without failures and array under test with failure elements. Fourier transform submatrix is used to constitute measurement matrix and a random undersampling strategy is implemented on the far field radiation patterns of reference array and array under test to acquire measurement data. Three ISTA, including Parallel Coordinate Decent (PCD) Algorithm, Separable Surrogate Functional (SSF) Algorithm, Iterative Reweighted Least Square (IRLS) Algorithm are then applied to reconstruct the sparse signal. Several simulation experiments are carried out and the results indicate that proposed method not only locates the positions of failure elements precisely, but also identifies their failure types using far less measurement points compared with traditional methods.