Successive interference cancellation receiver with neural network compensation in the CDMA systems

In code division multiple access (CDMA) systems, the multiple access interference (MAI) is one of the important factors that limit the system capacity. We propose a successive interference cancellation (SIC) scheme with neural network compensation to efficiently reduce the effect of MAI. In the literature, multi-stage SIC is often used to achieve an acceptable bit error rate (BER). Although multi-stage SIC can significantly reduce the BER, it will increase the system complexity as well. We propose a neural network-based multi-user SIC receiver that will both achieve low BER and reduce system complexity. The experimental results show our propose system has better BER performance than original SIC receiver at 30%.