A Simple Subspace-Based Semi-Blind Channel Estimator for Precoded OFDM Systems

Abstract

In this paper, a simple subspace based semi-blind channel estimator is developed for precoded OFDM systems. This channel estimator relies on redundant linear preceding. With this proposed channel estimator, the channel identifiability is guaranteed up to one scalar ambiguity, regardless of the channel zero locations and the underlying signal constellations. The ambiguity can be resolved with the known pilots inserted in the transmitted sequences. Computer simulations are conducted to support our analysis. The simulation results show that, compared with the well known LS estimator, the proposed estimator achieves higher estimation accuracy and is capable of tracking slow channel variations.