Matrix graduated nonconvexity annealed neural network for DS-CDMA multiuser detector

We propose an efficient two-stage DS-CDMA multiuser detector (MUD). The first stage is a reduced linear detector and the second stage is a matrix graduated nonconvexity annealed neural network (MGN-ANN). By using a first stage linear detector, the computational complexity in the second stage can be significantly reduced. We carried out extensive simulations in order to compare the error probability performance of our proposed detector with other competing multiuser detectors, which are based on conventional matched filter detector (CD), Hopfield neural network (HNN), multistage detector (MS-10) and annealed neural network (ANNMD). We also use the optimum multiuser detector (OMD) as comparison benchmark for all the MUDs. We show that the error probability performance of our proposed detector is significantly better than the other suboptimum MUD´s and approaches the performance of the OMD.