一种不依赖超参数的稀疏信号单快拍DOA估计方法

Hyperparameter-free sparse signal direction-of-arrival estimation method with single-snapshot

基于传感器阵列输出模型的稀疏重构,研究了利用单快拍数据进行相干信号波达方向(direction-of-arrival,DOA)估计的问题。定义一个干扰协方差矩阵作为权矩阵,通过加权最小二乘(weighted least squares,WLS)准则的迭代自适应求解,实现单快拍DOA高精度估计算法,简称WLS-IAE算法。详细分析了算法的计算复杂度,并与经典的稀疏估计类算法进行比较。结果表明作为一种稀疏表示类估计方法,WLS-IAE算法不仅保持了在低信噪比、单快拍、信号相干、信号DOA角度间隔小等非理想条件下的良好估计性能,而且无需选取超参数,计算复杂度更低,具有更强的实时性,适用于快变目标信号DOA的实时跟踪测量,具备潜在的工程实用价值。仿真实验验证了提出算法的有效性。

Based on the sparse representation of the array output model,the issue of direction-of-arrival(DOA)estimation for coherent signals is researched utilizing the single-snapshot data.An interference covariance matrix is defined as a weighed matrix,and the high-precision estimation of single-snapshot DOA is realized by the iterative adaptive solution of a weighted least squares(WLS),abbreviated to WLS-IAE.The computational complexity of the algorithm is analyzed in detail,and compared with that of the classical sparse estimation algorithm.As a sparse representation estimation method,the proposed algorithm not only maintains good estimation performance under non-ideal conditions such as low signal to noise ratio,single snapshot,coherent sources and nearby sources,but also do not require to select hyperparameters,and has low computational complexity and hard real time,which is particularly useful for the real-time measurement and tracking of time-varying DOA with potential practical value.Simulation experiments verify the efficacy of the proposed algorithm.