Noise Subspace Channel Estimation Algorithm for OFDM Systems

Abstract

In this paper, a novel blind channel estimator for Orthogonal Frequency Division Multiplexing (OFDM) affected by unknown impulsive interference is proposed. Unlike conventional subspace-based methods, this approach combines noise subspace decomposition with eigenvalue filtering to enhance interference suppression and improve channel estimation accuracy. The proposed technique ensures precise estimation of the covariance matrix, which is critical for reliable channel state information (CSI) retrieval. Moreover, by incorporating the presence of virtual subcarriers, the method further refines the estimation of channel response. Simulation results demonstrate that the proposed algorithm significantly outperforms existing subspace-based estimators, particularly in highly time-varying wireless channels and low signal-to-noise ratio (SNR) conditions. These findings confirm the practical applicability of the method in next-generation wireless networks like 5G and beyond, where robust and accurate channel estimation is essential for maintaining communication reliability under adverse conditions.