Publication

Abstract : Intelligent reflecting surface (IRS) assisted wireless communication is a potential solution for future wireless networks. With minimal hardware cost, it improves the symbol error rate (SER), throughput, and energy efficiency of the wireless system. In this paper, we investigate the problem of joint carrier frequency offset (CFO) and channel estimation in an IRS-assisted orthogonal frequency division multiple access (OFDMA) uplink system. We develop a low-complexity joint estimation method utilizing the Bayesian framework by assessing the sparse nature of the wireless channel. In a sparse Bayesian learning (SBL) framework, the CFOs and channels are jointly estimated iteratively using the space-alternating generalised expectation-maximization (SAGE) method. We derive the Cramer-Rao Lower Bound (CRLB) of the joint estimation problem to benchmark the performance of the proposed technique. The normalized mean squared error (NMSE) is used to evaluate the performance of the proposed estimation method. Furthermore, we propose an optimization technique for the reflection coefficients by maximizing the channel gain of the dominant line-of-sight (LoS) tap in the time domain.