一种新的单次快拍二维ESPRIT算法

针对复杂电磁干扰背景下相干信源的二维波达方向快速估计问题,从减小协方差矩阵计算量角度,提出了一种新的单次快拍二维ESPRIT算法(SS-ESPRIT). 该算法仅用一次快拍数据构造4个等效的协方差矩阵,进一步构造扩展的等效协方差矩阵,通过对其一次特征分解,即可实现完全解相干和二维波达方向估计. 为进一步提升该算法估计性能,提出了同相位数据叠加的对策. 数值仿真验证了SS-ESPRIT算法在提升实时性的同时,不会造成估计性能的下降,仅利用一次快拍数据的该算法估计性能优于快拍数为50次的空域平滑波达方向矩阵算法(DOAM),且接近快拍数为100次的DOAM算法,叠加8次同相数据后的该算法性能明显优于200次快拍的DOAM算法. 结果表明新算法适用于小数据样本估计或对实时性要求高的应用背景. 

A Novel Single Snapshot Two-Dimensional ESPRIT Algorithm

In order to estimate two-dimensional (2-D) direction-of-arrival(DOA) of coherent signal sources fast in complex electromagnetic interference environments, a novel two-dimensional ESPRIT method called single snapshot ESPRIT algorithm(SS-ESPRIT) is proposed. The proposed algorithm reconstructs four equivalent covariance matrices using only one snapshot data from arrays. The four equivalent covariance matrices compose an extended equivalent covariance matrix. De-correlation and 2-D DOA estimation could be accomplished via the eigen-decomposition of the matrix. The strategy of in-phased data accumulation is also proposed to improve the performance of SS-ESPRIT algorithm. Simulation results show that SS-ESPRIT algorithm could realize fast estimation of 2-D DOA without causing the performance degradation. The estimation performance of new algorithm using single snapshot is better than that of direction-of-arrival matrix (DOAM) algorithm using 50 snapshots and is close to the performance of DOAM algorithm using 100 snapshots. The estimation performance of the algorithm using 8 in-phased data is better than that of DOAM algorithm using 200 snapshots. It could be concluded that the algorithm proposed in this research is suitable for application in short-time sampled data and real-time estimation.