An Underwater Neural Network DOA Estimation Model with Fast Convergence and Strong Robustness

Due to the complexity and variability of the underwater environment, DOA estimation algorithms based on mathematical models will produce errors or even fail. In addition, the neural network has the ability of generalization and mapping. It can consider the noise, transmission channel inconsistency, and other factors of the objective environment. Therefore, this paper utilizes the Back Propagation (BP) neural network as the basic framework of underwater DOA estimation. In addition, in order to improve the DOA estimation performance of the traditional BP neural network, multi-source underwater DOA estimation of PSO-BP-NN based on a High-order Cumulant optimization algorithm are proposed. Finally, the effectiveness of the proposed algorithm is proved by comparing it with the state-of-the-art algorithms and MUSIC algorithm.