Abstract: To solve the problem that impulsive noise interference deteriorates the system link reliability of orthogonal frequency division multiplexing (OFDM) systems, the turbo iterative equalization detection method based on Markov chain Monte Carlo (MCMC) is proposed. Firstly, the transmitter and receiver models of the OFDM system are given. Then, the prior distribution of unknown variables and their conditional distribution given the received signal are derived. The parameters of channel and impulse noise can be estimated by samples generated from the MCMC equalizer. Finally, specific implementation processes of turbo iterative structures are provided. Simulation results show that the proposed method can reduce the impact of impulsive noise and significantly improve the system reliability in additive white Gaussian noise (AWGN) channels and multipath channels.