Semi-Blind Signal Estimation Using Toeplitz Structure-Based Subspace Method

Abstract

This paper presents a novel semi-blind signal estimation method for a multi-channel system based on a structure-based subspace technique. The proposed method exploits the Toeplitz structure present in the signal matrix of the communication system model and the presence of some pilot sequences. The use of pilot symbols improves the efficiency and robustness of the algorithm and also eliminates the scalar ambiguity inherently present in blind systems. The devised method estimates the transmitted symbols directly without having to estimate the channel as an intermediate step. This will significantly reduce the error propagation effect and eventually results in an improved accuracy. Simulation results illustrate that the developed method exhibits an attractive performance.