子空间跟踪PASTd算法的改进及其应用

分析了PASTd（紧缩近似投影子空间跟踪）算法，通过仿真分析发现PASTd算法在信噪比较高的情况下，特征值收敛速度达不到MDL（最短描述长度）算法的要求，导致维数估算的结果偏大。针对这一问题，文中把OPAST（正交近似投影子空间跟踪）算法引申到PASTd算法中，弥补它的缺陷。仿真结果表明，提出的改进算法收敛性能较原PASTd要好，可得到更好的维数估计与波达方向估计。另外文中提出的算法对天线阵列的干扰零点形成有很强的实用价值。

Improvement and Applications of PASTd Algorithm for Subspace Tracking

The PASTd algorithm is analyzed.By simulation it is found that when the SNR is relatively high,the convergence speed of signal eigenvalues can not meet the demand of MDL,resulting in higher rank estimation.In this paper a Fast Orthonormal PAST Algorithm(OPAST) is introduced into the PASTd algorithm for tackling this shortcoming.It is shown by simulation that the improved algorithm can achieve better convergence and better estimation of the rank and DOA.So it is very useful for nulling the interference.