基于凸优化的Khatri-Rao子空间宽带DOA估计

Khatri-Rao Subspace Wideband DOA Estimation Based on Convex Optimization

该文基于凸优化工具,提出一种新的Khatri-Rao子空间宽带到达角(DOA)估计方法。首先,利用Khatri-Rao子空间的概念,构造各频点的虚拟阵列导向矢量。再利用凸优化方法针对虚拟阵列导向矢量构造聚焦矩阵,从而利用虚拟阵列所增加的维数,在尽量减少聚焦对噪声影响的同时,获得良好的聚焦效果。仿真结果表明,相对于已有的Khatri-Rao子空间宽带DOA估计方法 FKR-RSS,该方法具有更好的估计精度和目标分辨力。对于信号源数大于阵元数的情形,该文方法优势更加明显。

A convex optimization based Khatri-Rao subspace wideband Direction-Of-Arrive （DOA） estimation algorithm is proposed in this paper. First of all, the steering vectors of the virtual array are formulated with the Khatri-Rao subspace. And then these steering vectors of the virtual array are used to compute the wideba^d focusing matrix with convex optimization. Thanks to the increased dimensions of the virtual array, the proposed focusing matrix can minimize the distortion of the noise caused by the focusing procedure and maintain the good focusing performance as well. Simulations show that the proposed method performs better than the conventional Khatri-Rao subspace wideband DOA estimation algorithm： Focusing Khatri-Rao subspace-Rotational Signal Subspace （FKR-RSS） and when the number of sources is more than the number of sensors, the proposed method has more obvious superiority especially.