Single Snapshot DOA Estimation Based on Learned Modified Smoothed L0 Algorithm
Hangui Zhu, Cunqian Feng, Weike Feng, Chengliang Liu · 2021 CIE International Conference on Radar (Radar) · 2021
Based on the sparsity of signal, sparse recovery (SR) method can use single snapshot data for high-resolution Direction of Arrival (DOA) estimation of correlated signal sources. However, typical SR methods suffer from the problems of parameter setting difficulty, high computational complexity, and low recovery accuracy, limiting their practical applications. To solve these problems, based on the theory of intelligent learning, this paper unfolds the Modified Smoothed LO (MSLO) algorithm into a deep network, proposing the Learned MSLO (LMSLO) algorithm. By developing the deep LMSLO network and based on a complete training dataset, the proposed method can learn the best parameters of the MSLO algorithm, e.g., the approximation parameter, the iteration step, and the steepness factor, hence improving the convergence performance of the MSLO algorithm. Simulation results show that the proposed algorithm can obtain higher DOA estimation performance with lower computing cost than the SLO and MSLO algorithms.