为了解决相干信源的测向与跟踪问题,提出了一种基于阵列虚拟平移的快速解相干幂迭代算法。算法把阵列虚拟平移和幂迭代方法相结合,对数据协方差矩阵进行了重构,在不损失阵列孔径的情况下较好地实现了解相干处理,使直线阵可估计的相干信源数目达到了M-1个。避免了特征值分解和一维谱峰搜索,算法的收敛速度快,运算量小,算法同样适用于均匀圆阵。通过仿真实验,验证了算法的有效性。
For the need of coherent signal source bearing and tracking,a fast de-correlation power iterative subspace tracking algorithm based on array virtually removing is put forward,combining the array virtually removing with the power iterative method together.The algorithm reconstructs the data covariance matrix,achieves de-correlation without array aperture loss and improves the source number up to M-1,and also avoids EVD and spectrum-peak searching process.The propused algorith has a fast convergent speed and small computation burden,can meet the need for solving rank-1 subspace estimation and tracking.Furthermore,this algorithm can also be applied in UCA.The simulation experiment results show that this algorithm is effective.