Joint Estimation of Channel and Impulsive Noise Based on Maximum Correntropy Criterion Kalman Filter for NOMA System

The presence of impulsive noise (IN) significantly degrades the performance of non-Orthogonal multiple access (NOMA) systems. This letter proposes a novel maximum correntropy criterion (MCC)-based Kalman filter and smoother (KFS) algorithm for joint channel and IN estimation by exploiting all sub-carriers. In specific, we firstly formulate the channel and IN as a joint unknown sparse vector. Then, by exploring the statistical characterization of IN, we incorporate MCC with KFS to improve the robustness of KFS against IN. Finally, the expectation maximization (EM)-based sparse Bayesian learning (SBL) is combined with the MCC-KFS algorithm to realize the iterative estimation of the unknown hyperparameters. Simulation results validate the effectiveness of the proposed algorithm in channel estimation accuracy and bit error rate (BER) performance of NOMA systems under IN environments.