基于相干积累矩阵重构的波达方向估计新方法

Novel Direction Of Arrival Estimation Method Based on Coherent Accumulation Matrix Reconstruction

针对短时小样本条件下相干信号的波达方向(Direction Of Arrival,DOA)估计问题,该文提出了一种基于相干积累矩阵重构的快速解相干方法。首先利用相干积累技术对阵列接收快拍进行处理,得到累积快拍矢量,提高了数据信噪比。再依据累积快拍矢量的结构特点构造一个非降维等效协方差矩阵,理论分析可知,该矩阵的秩仅与信源个数相等,与信号间相关性无关,即实现了相干信源完全解相干。相较于空间平滑类算法,该方法避免了阵列孔径损失,估计精度高、计算量小。仿真结果验证了算法的有效性。

Based on coherent accumulation matrix reconstruction, a novel Direction Of Arrival （DOA） estimation decorrelation method of coherent signals is proposed using a small sample. First, the Signal to Noise Ratio （SNR） is improved by performing coherent accumulation operation on an array of observed data. Then, according to the structure characteristics of the accumulated snapshot vector, the equivalent covariance matrix, whose rank is the same as the number of array elements, is constructed. The rank of this matrix is proved to be determined just by the number of incident signals, which realize the decorrelation of coherent signals. Compared with spatial smoothing method, the proposed method performs better by effectively avoiding aperture loss with high-resolution characteristics and low computational complexity. Simulation results demonstrate the efficiency of the proposed method.