Compressed Sensing-Based DOA Estimation With Antenna Phase Errors

In array signal processing, the direction of arrivals (DOAs) of the received signals are estimatedby measuring the relative phases among antennas; hence, the estimation performance is reducedby the inconsistency among antennas. In this paper, the DOA estimation problem of the uniformlinear array (ULA) is investigated in the scenario with phase errors among the antennas, and adiagonal matrix composed of phase errors is used to formulate the system model. Then, by using thecompressed sensing (CS) theory, we convert the DOA estimation problem into a sparse reconstructionproblem. A novel reconstruction method is proposed to estimate both the DOA and the unknownphase errors, iteratively. The phase errors are calculated by a gradient descent method with thetheoretical expressions. Simulation results show that the proposed method is cost-efficient andoutperforms state-of-the-art methods regarding the DOA estimation with unknown phase errors.