Phase Noise Estimation and Mitigation for Cognitive OFDM Systems

In this paper, a novel maximum likelihood (ML) method is proposed to estimate and mitigate phase noise (PN) for cognitive orthogonal frequency division multiplexing (OFDM) systems. In the proposed method, PN estimation is formulated as a unitary-constrained optimization problem based on the ML criterion. Using the obtained estimate of the PN vector, both common phase error and inter-carrier interference are effectively mitigated and reduced for symbol error rate (SER) performance improvement. Simulation results show that the proposed method can mitigate PN effectively, and obtain better SER performance for cognitive OFDM systems compared with conventional methods.