FDPM aided decision directed channel estimation with VSSNLMS-based predictor for OFDM systems

Abstract

The next generation broadband wireless communication system will largely depend on high rate transmission scheme such as Orthogonal Frequency Division Multiplexing (OFDM) scheme. However, channel estimation technique remains a central factor to efficient performance of the OFDM transmission scheme in wireless communication systems. A Fast Data Projection Method (FDPM) - based Channel Impulse Response (CIR) estimator and a Variable Step Size Normalized Least Mean Square (VSSNLMS)-based predictor are proposed for Decision Directed Channel Estimation (DDCE) scheme designed for OFDM system. These algorithms are employed to implement the DDCE scheme by considering a more realistic channel condition, the Fractionally Spaced (FS)-CIR-based channel model. The performance of the FDPM based CIR estimator with VSSNLMS predictor is compared with the case where NLMS predictor is employed. The performance of the formal is observed to improve significantly over the later. The FDPM aided DDCE scheme is also compared with an earlier proposed deflated Projection Approximation Subspace Tracking (PASTd) algorithm aided DDCE scheme. It was found from the simulation results that the FDPM-based DDCE exhibits an improved performance over the PASTd-based DDCE.