无需信源数目的改进TOFS算法  

Modified TOFS Algorithm with Unknown Number of Sources

作　　者：曾耀平 ZENG Yaoping(School of Communication and Information Engineering,Xi＇an University of Posts and Telecommunications,Xi＇an 710121,China)

机构地区：[1]西安邮电大学通信与信息工程学院,西安710121

出　　处：《微处理机》2018年第3期43-46,共4页Microprocessors

基　　金：陕西省教育厅专项基金项目(17JK0693)

摘　　要：针对投影子空间正交类宽带DOA估计算法性能不够稳健且需要信源数目预估的问题,提出了一种无需信源数目预估的宽带信号DOA估计方法。该方法首先对阵列协方差矩阵进行特征分解获得对应的特征向量,然后对带宽范围内所有频点的特征向量进行全空间加权,无需信源数目就可以得到估计的阵列噪声子空间。结合TOFS算法,既克服了TOPS算法依赖选择的参考频点进而造成其性能不够稳定的缺点,同时也消除了TOPS算法容易出现伪峰的难题。新算法无需进行信源预估就可以实现宽带信号的高分辨角度估计。计算机仿真结果表明所提算法的精度较高、运算量适中,具有较高的稳健性。In order to solve the problem that the performance of orthogonal wideband DOA estimation algorithm in projection subspace is not robust enough and the estimation of source number is needed, a wideband DOA estimation method without source number estimation is proposed. The method firstly decomposes the covariance matrix of the array to obtain corresponding eigenvectors, and then weights the eigenvectors of all frequency points in the bandwidth in full space, so that the estimated subspace of array noise can be obtained without the number of sources. Combined with TOFS algorithm,it not only overcomes the shortcoming that TOPS algorithm relies on the selected reference frequency point and thus its performance is not stable, but also eliminates the difficulty that TOPS algorithm is prone to false peaks. The new algorithm can realize high resolution angle estimation of wideband signals without source estimation. The computer simulation results show that the proposed algorithm has high precision, moderate computation and high robustness.

关 键 词：宽带信号 DOA估计 无需信源数目 TOFS算法 

分 类 号：TN911.7[电子电信—通信与信息系统]