A Novel Neural Network Blind Multi-user Detection Algorithm

Blind multi-user detection (BMUD) is a key technology in CDMA to improve communication quality. This paper introduced FNN (feed-forward neural network) to BMUD algorithm. It combined FNN with CMA algorithm, constructed a cost function, optimized FNN weights and parameters by LMS and then realized BMUD. Compared with traditional CMA blind multi-user algorithm, simulation results indicate new algorithm improves the performances in bit-error ratio, following ability and soon.