Estimation of Signal Parameters Using SSA and Unitary Root-Music

Abstract

The improving of performance of the signal parameter estimation when using SSA approach is obtained in the paper. The unitary transformation of extended data matrix obtained after SSA technique or corresponding covariance matrix is used. Such transformation reduces the computational load and improves the performance of subspace-based techniques used with SSA approach. This performance improvement is due to the so-called forward–backward averaging effect. The reduction of computational load is mainly due to the fact that the real-valued eigendecomposition instead of complex one is performed. Examples of applications of proposed approach in the modern communication systems and the ways of future investigations are considered.