Neural network techniques for the optimal DS/CDMA multi-user detector

摘要

In this paper, the application of the mean field annealing neural network to the multi-user detection in DS/CDMA systems is proposed. It is shown that the NP-complete problem of the optimal multi-user detection can be translated into minimizing an MFA network energy function. Numerical results show that the proposed detector offers significant performance gain relative to the conventional detector and decorrelating detector while it can employ neural networks and be implemented easily in analog hardware.