Multi-user detection using non-parametric Bayesian estimation by feed forward neural networks

This paper is concerned with developing novel encoding techniques for implementing non-parametric neural based detectors for systems using Code Division Multiple Access. These new encoding methods on the one hand can increase the processing speed and reduce the complexity of the Feed Forward Neural Network based detector, on the other. Furthermore, we demonstrate that an asymptotically optimal detection performance can be achieved by the proposed algorithms. Due to the increased processing rate, the new scheme may further improve Spectral Efficiency. Extensive simulations and the corresponding numerical analysis demonstrate that the proposed algorithms yield near optimal performance on real channel models (COST-207).