基于交错投影的无网格降维DOA估计算法

Dimension-reduced Grid-less DOA Estimation Algorithm Based on Alternating Projection

针对当前基于交替投影的无网格(APG)DOA估计算法存在收敛速度较慢的问题,提出了一种基于交错投影的无网格降维DOA估计算法。该算法在迭代求解的过程中,直接对半正定矩阵的子矩阵——Toeplitz矩阵进行迭代投影,降低了矩阵的维度,在确保估计精度的情况下,有效降低了算法的计算量。仿真结果表明该算法与APG算法相比,在相同信噪比、快拍数或阵元数的条件下能够保证估计精度,同时减少了计算时间,具有更好的实时性,性能更优。

In view of the problem of slow convergence of current Grid-less DOA estimation algorithms based on alternating projection,a grid-less dimensionality reduction DOA estimation algorithm based on staggered projection was proposed.In the process of iterative solution,the iterative projection of the submatrix of the positive semi-definite matrix,the Toeplitz matrix,was directly performed,which reduced the dimension of the matrix,and effectively reduced the calculation amount of the algorithm while ensuring the estimation accuracy.Simulation results showed that,compared with the APG algorithm,the algorithm could guarantee the estimation accuracy under the same signal-to-noise ratio,snapshot number or number of elements,while reducing the calculation time,and had better real-time performance and better performance.