基于人工鱼群算法的声矢量传感器阵列的最大似然DOA估计

Maximum Likelihood DOA Estimation Based on Artificial Fish Swarm Algorithms for Acoustic Vector Sensor Array

在阵列信号处理中,极大似然法(ML)对波达方向(DOA)估计有很好的性能,但由于多维非线性搜索的计算复杂,很难应用于工程中。为了降低ML方法的计算复杂度,提出了一种改进人工鱼群算法(AFSA)的声矢量传感器阵列的ML-DOA估计方法。仿真结果表明,与基于遗传算法(GA)、粒子群优化算法(PSO)和微分进化算法(DE)的ML-DOA估计相比,该算法具有更快的收敛速度、更低的RMSE、更低的计算复杂度和更稳定的性能。

Maximum likelihood(ML)method has good performance for direction of arrival(DOA)estimation in array signal processing,but it is hardly applicable to engineering because of the computation complexity in determining the signal azimuth by multi-dimensional nonlinear search.In order to reduce the computational complexity of ML method,an ML DOA estimation based on Artificial Fish Swarm Algorithms(AFSA)algorithm for acoustic vector sensor array was proposed.The simulation results show that proposed algorithm has faster convergence speed,lower RMSE,lower computational complexity,and more stable performance compared with the ML DOA estimation based on genetic algorithm(GA),particle swarm optimization(PSO),and differential evolution(DE)algorithm,and it is more suitable for engineering applications.