Efficient DOA Estimation Algorithm Based on Unbiased Variable-Step-Size LMS Algorithm

DOA estimation algorithm based on adaptive filtering divided the received signal into desired signal and auxiliary signal, and solved the weight parameters through adaptive structure iteration, making the weighted auxiliary signal approach to the desired signal, so as to form the spatial spectrum peak in the direction of signal arrival. Traditional fixed-step-size LMS algorithm is used to solve the weight parameters, but it is difficult to choose a suitable step size. The variable-step-size LMS algorithm can adjust the step size according to SNR, but it is still a biased estimation of the weight parameters in principle. In this paper, an unbiased variable-step-size LMS algorithm is proposed to compensate the bias on the loss function so as to realize the unbiased estimation of the weight parameters, and further improve the performance of DOA estimation. The feasibility of the algorithm is verified through simulation experiments, and the performance of the proposed algorithm is compared with that of the traditional algorithm.