A hybrid digital signal processing-neural network CDMA multiuserdetection scheme

摘要

We present a new hybrid digital signal processing-neural networkntwo-step multiuser detection scheme whose small computational complexitynmakes it attractive for real-time CDMA multiuser detection. Anninvestigation on the nature of the local minima of the Optimal MultiusernDetector's (OMD) objective function leads to the development of annefficient algorithmic stage that can reduce significantly the size ofnthe OMD optimization problem. This stage may then be followed by anHopfield neural network employed to solve a smaller size residualnproblem of the same form. The performance of the hybrid detector isnevaluated via simulations and it is shown to exceed that of othernsuboptimal receivers at a much lower computational cost in bothnsynchronous and asynchronous CDMA transmission cases