DOA estimation based on least squares fitting with sensor gain and phase uncertainties for Doppler signals

We propose several preprocessing techniques that improve the performance of the LSF based method for DOA estimation in the presence of rank deficiency in time correlation matrix, which frequently occurred in practical applications. In these given methods, the degrees of freedom are reduced with several novel schemes being taken to fulfill the ill conditioned optimization problem. The given techniques include temporal smoothing, subspace constraint, DFT transformation and diagonal loading. Monte Carlo simulation results verify the efficacy of the proposed modifications.