Perceptron Algorithm for Channel Shortening in OFDM System with Multipath Fading Channels

Channel shortening methods, in multicarrier systems, are applied for decreasing and almost compensating for the inter-symbol and inter-carrier interferences due to the channel delay spread. In this paper, we propose a new channel shortening technique for orthogonal frequency division multiplexing (OFDM) systems based. The propoed method is based on the neural network equipped with the Perceptron learning rule. Also we have tested our method in the OFDM system with multipath fading channels. The simulation results and mathematical analysis show the better performance of the proposed method, with BER criterion, compared to the commonly used channel shortening methods such as MMSE, MSSNR and MERRY in multicarrier systems. Also the proposed method has almost the same computational complexity, as well as the mentioned methods.