Ball’s-Based Adaptive Channel Estimation Scheme Using RLS Family-Types Algorithms

Abstract

The recovery of the transmitted signals, propagated through wireless communication channels, is a complicated process due to the distortion and interference impairments in such random channels. However, the channel effect can be compensated by using channel estimation techniques performed at the receiver. Since most of channel estimators are operated in the frequency domain, i.e., inverse modeling, the Ball’s adaptive channel estimation scheme, which was invented by Michael J. Ball, was not considered in the literature as an attractive approach due to its time-based and direct modeling features. However, such features are favorable for time-varying channels. To the best of author’s knowledge, the performance of the Ball’s scheme for channel estimation has not been investigated in the context of OFDM system with fading channels. Therefore, in this paper, we propose to incorporate Ball’s method for channel estimation in OFDM receivers and we also investigate the performance of this scheme using variations of the Recursive Least Squares (RLS)-type algorithms, namely QR Decomposition RLS (QRD-RLS), the Householder RLS (HH-RLS) and the Sliding Window Householder RLS (SWHH-RLS). Our numerical results indicate that QRD-RLS and HH-RLS outperform traditional RLS in terms of bit error rate.