Blind CFO Estimation for Linearly Precoded OFDMA Uplink

This paper presents a novel method for blind carrier-frequency-offset (CFO) estimation in linearly precoded OFDMA uplink. Our investigation starts from the single-user equivalent (SUE) scenario presented in where active users in the network are categorized into a number of reference (synchronized) users (RUs) and a new (asynchronous) user (NU). The major idea is to take advantage of time correlation induced by the linear precoder, which offers a second-order moments-based blind CFO estimation. The precoder design is carefully performed in terms of CFO identifiability, estimation accuracy and overall system performance. In the multiuser scenario, where all users can be misaligned in frequency domain, the proposed CFO estimator is capable of mitigating a considerable portion of interference from neighboring users through exploitation of a novel time-frequency multiuser data-mapping (MU-DM) scheme. To demonstrate the multiuser interference (MUI)-resilience feature of proposed scheme, theoretical analysis is performed through derivation of approximate minimum-mean-square error (MMSE) in both SUE and multiuser scenarios. It is shown that by exploitation of the proposed MU-DM scheme, the approximate multiuser MMSE is very close to that of SUE scenario. Simulation results show that the proposed approach outperforms state-of-the-art approaches in both SUE and multiuser scenarios.