针对单次快拍情况下相干信源波达方向(DOA)估计的解相干和实时性问题,提出了基于空间平滑的单快拍DOA估计算法。该算法直接采用单次快拍数据的前后向空间平滑子阵,构造centro-Hermitian伪协方差矩阵,实现解相干,而无需对入射信号的初始相位作特别假定,也无需计算各个子阵的协方差矩阵,减小了计算量;利用酉变换将复数域的奇异值分解实数化,降低了计算复杂度;结合MUSIC、ESPRIT等子空间类算法,实现了相干信源的DOA估计。数值仿真表明该算法无需对入射信号的初始相位作特别假定,且对于相干信源能够有效实现单快拍DOA的快速估计,适用于对实时性要求较高的场合。
A direction-of-arrival (DOA) estimation algorithm based on spatial smoothing was proposed for the decorrelation and instant problems among DOA estimation of coherent signals when there was only one valid snapshot. The forward/backward spatial smoothing subarrays of single snapshot were directly used to construct the pseudocovariance matrix to accomplish decorrelation, which does not require that signals' natural phases must be zero, and that the covariance of every subarray must be computed. The pseudocovariance, which was a centro-Hermitian matrix, was changed into real one by unitary transformation to reduce the computational complexity. Combined with the subspace kind algorithms such as MUSIC and ESPRIT, the DOA ccould be estimated. Simulation results showed that the algorithm did not require special assumption on signal's natural phases, and could estimate the single snapshot DOAs of coherent signals effectively, which was suitable for real-time demand higher occasion.