Channel Parameter Estimation in the Presence of Phase Noise Based on Maximum Correntropy Criterion

Phase noise (PN) is a prevalent challenge in oscillator-driven systems, leading to spectral dispersion of the power spectral density (PSD) around a Dirac delta function. This paper addresses the task of estimating a communication channel affected by additive white Gaussian noise (AWGN) and phase noise. Traditional estimation methods such as the least mean square (LMS) and mean square error (MSE) criteria are deemed inadequate due to the unique characteristics of phase noise. In this study, we propose a novel approach for PN channel estimation utilizing information-theoretic learning (ITL) principles, specifically focusing on the maximum correntropy criterion (MCC). By employing MCC, our method enhances the robustness of the channel estimator in steady-state conditions, thereby improving the accuracy of parameter estimation. Additionally, to expedite the convergence rate of our algorithm, we introduce a novel mixed-LMS approach that amalgamates elements of both MSE and MCC. This hybrid technique leverages the strengths of each criterion, resulting in a more efficient and accurate estimation of the PN-affected channel. Through comprehensive analysis and experimentation, our proposed method demonstrates its effectiveness in mitigating the impact of phase noise on channel estimation.