基于引力搜索算法的相干信号源DOA估计方法

针对最大似然(ML)DOA估计方法存在着运算量高且容易收敛到局部极值的问题。结合引力搜索算法(GSA)与最大似然方法,提出了一种GSA-ML方法。将最大似然函数作为GSA算法的适应度函数,在遵循ML方法的主体思想同时,利用GSA算法运算量低和收敛速度快的优点,成功地找到似然函数的全局最优解;并保存了ML方法的优点。仿真结果表明,GSAML方法不仅能有效估计相干信号源;并且相比MUSIC、ESPRIT和TLS-ESPRIT算法,拥有更高的精度和估计成功概率。

DOA Estimation Method of Coherent Signal Sources Based on Gravitational Search Algorithm

For the maximum likelihood (ML) DOA estimation method, there exists high computational complexity and easy convergence to the local extremum. This paper combines the Gravitational Search Algorithm(GSA) and the maximum likelihood method to propose a GSA-ML method. This method uses the maximum likelihood function as the fitness function of GSA algorithm, under the main idea of ML method, and utilizes the low computational cost and fast convergence rate of GSA algorithm, to successfully find the global optimal solution of likelihood function and save the advantages of ML method. Simulation shows that the GSA-ML method proposed in this paper not only can effectively estimate the coherent signal source, but also has higher accuracy and estimated probability of success than the MUSIC,ESPRIT and TLS-ESPRIT algorithm.