Channel Estimation Based on Neural Network in OFDM System for Powerline Communication

In this paper, channel estimation based on neural network in powerline communication is proposed and its performance is compared with LS and MMSE methods by computer simulations using mean square error (MSE) and bit error rate (BER) criterias. The MSE and BER performances of neural network to estimate channel are between the LS and MMSE algorithms. MMSE algorithm yields the best performance but its complexity is high. The advantage of the use of the neural network is that the neural network yields better performance than the LS algorithm and it is less complex than the MMSE algorithm(1).