Interference cancellation in DS-CDMA systems using neural network techniques

The capacity of Direct Sequence-Code Division Multiple Access (DS-CDMA) systems is limited by the Multiple Access Interference (MAI). MAI is produced due to the unorthogonality of the used codes. To improve the system performance, different MAI cancellation techniques have been proposed. These techniques introduce some performance improvements over the conventional receivers. This paper presents a Neural Network (NN) based receiver that has the capability to reject the MAI and outperforms the conventional matched filter receiver. Also, this receiver can combat the 'near-far' problem. A comparative performance analysis of the conventional matched filter receiver and the NN receiver is carried out using computer simulation. It is found that the new receiver performs extremely well in MAI and in 'near-far' condition as well.