An intelligent receiver for space time block coded OFDM system

摘要

An intelligent receiver based on Radial Basis Function neural network (RBF) for Space-Time Block Coded MIMO system is developed here. Due to their fast maximum-likelihood(ML) decoding, STBC are commonly used in MIMO-OFDM system. However, most of the existing STBC methods rely on the availability of accurate channel state information (CSI) at the receiver. Furthermore, many STBC assume the channels are block-wise constant, but these do not always hold. So RBF based receiver is designed in the paper. With respect to the RBF neural network, PCA(Principal Component Analysis) is applied to process the train samples and growing and pruning(GAP) algorithm is adopted as the learning algorithm. Compared with ML decoder, the proposed receiver has a high precision and a good performance to track the variations of the fading channels. The result of simulation illustrates the effectiveness of the receiver introduced.