Fast subspace estimation with Rayleigh-Ritz approximation and multi-stage Wiener filter

In this paper, the Rayleigh-Ritz approximation and the forward recursions of the multi-stage Wiener filter are adopted to perform the fast estimation of signal subspace, which does not need the estimation and eigen-decomposition of the covariance matrix, and is, therefore. computationally advantageous. Moreover, the rigorous form of the desired signal is not required. By complexity analysis and numerical simulation it is concluded that the proposed method can achieve better performance with lower computational cost, especially in the case of lower signal-to-noise ratio.