An adaptive Maximum A-Posteriori based channel estimation method is presented in this paper. The estimator is proposed to be used in Orthogonal Frequency Division Multiplexing systems in Power Line Communication. The proposed estimator estimates the subchannel correlation adaptively where the channel is assumed to be frequency selective and slowly time varying. The performance of the estimator is investigated by evaluating the Mean Square Error of estimations. The simulations are done using measured channel data taken from the literature with additive white Gaussian noise assumption. The results show that the proposed estimator has better performance compared to Maximum Likelihood estimator. This performance improvement is clearer especially for low to moderate SNRs.