A Hopfield Neural Network Based MIMO System

In this paper, we present a MIMO (Multiple Input Multiple Output) communication system that uses the Alamouti algorithm for transmission with an appended Hopfield Encoder at the transmitter and Hopfield Decoder at the receiver. The Hopfield Encoders and Decoders are a part of the Hopfield Neural Network that can be used as an associative memory. At the transmitter, the symbols to be transmitted are first separated into blocks of length say k. These blocks are then elongated in size to a length l = (2k/0.15). These blocks of data each of length l are then encoded using the Alamouti algorithm and transmitted over a Rayleigh fading channel. At the receiver, the received blocks are sent to a Hopfield network which acts as an associative memory and tries to retrieve one of the possible transmitted symbols. After retrieval, the original data symbols are decoded. It was observed that the performance of the system was enhanced by use of the Hopfield network compared to Alamouti system without Hopfield neural network.