基于短快拍的捷变频信号DOA估计算法

DOA Estimation Algorithm for Frequency Agile Signal Based on Short Snapshots

传统的波达方向(DOA)估计算法依赖于多快拍数据得到的协方差矩阵,在实际应用中,可用的快拍数并不确定,而捷变频信号由于其频率的不固定导致短快拍下的DOA估计更加困难.针对这一问题,提出一种基于短快拍的捷变频信号DOA算法,即ISSFAS(improved-short-snapshot-frequency-agile-signal)算法.该算法将伪协方差矩阵法与频域聚焦法相结合,用重新构成的伪协方差矩阵代替原有的协方差矩阵,并用频域聚焦的方法对可变频率的情况进行处理.研究结果表明:ISSFAS算法具有良好的DOA估计性能,能够在短快拍下进行超分辨,且性能强于未与伪协方差法相结合的情况.

The classical DOA estimation algorithm is based on the covariance matrix which is built from a lot of snap- shots. However, the number of the available snapshots is indeterminate in the practical application, and the DOA estimation of frequency agile signal is even more difficult because of the changing frequency. In order to solve the problem, a DOA estimation algorithm was proposed for frequency agile signal based on short snapshots, which is called ISSFAS （improved-short-snapshot-frequency-agile-signal）. This algorithm combines the method of constructing pseudo-covariance matrix and the method of frequency domain focusing, and it uses the pseudo covariance matrix which is newly built to replace the covariance matrix and uses the method of frequency domain focusing to deal with the problem of changing frequency. The research results show that the ISSFAS algorithm has good performance of DOA estimation, and it can super resolute in the case of short snapshots. Meanwhile, it outperforms the algorithm without pseudo-covariance matrix.