A GIVENS Based URV Subspace Method

n many modern subspacebased algorithms,the eigenvalue decomposition or the SVD is expensive to compute,which severely limits the realtime applicability of these algorithms.URV derives another set of base for noise subspace by relaxation the requirment of the diagonal matrix in SVD to upper triangular matrix,which reduces the cost of computation.URV can be easily expanded to subspace traking algorithm.This paper provides an URV decomposition based subspace realization method,which is carried out by GIVENS rotation.This method has modular structures and hence is suitable for hardware implementation.At the end of this paper,simulation results are given to demonstrate the effectiveness of the method.