基于 EM算法的宽带信号 DOA 估计及盲分离

摘要

本文提出了一种基于 EM算法的宽带信号 DOA 估计与盲分离方法。首先将宽带混合信号转换到频域，然后综合利用带宽内所有频点信息建立似然函数，在此基础上推导出宽带条件下的 EM迭代式，从而实现宽带信号DOA 及波形的联合估计。并且本文通过分析 EM算法的收敛性，自适应的设定角度搜索空间，提高了算法的运算效率。与传统方法相比，本文方法运用的有效信息更多，因此，其在 DOA 估计精度及波形恢复性能方面都更有优势。仿真实验表明了该算法的有效性。%Based on the expectation-maximization (EM)algorithm,a new method for broadband direction of arrival (DOA) estimation and blind source separation is presented in this paper.First,we establish a likelihood function in the frequency domain by jointly using all frequencies information.Then,based on this function,the E-step and M-step in the frame of broadband EM algo-rithm are derived.Finally,we jointly estimate DOAs and sources through EM iteration compute.Moreover,through analyses the con-vergence of EM algorithm,we set adapted angle search space for DOA estimation,which reduce the computations of our algorithm. Compare to the traditional method,more effective information are applied in our algorithm.Therefore,the proposed algorithm has more advantages for DOA estimation and source separation.Simulation results demonstrate our algorithm is effective.