Multiple access interference and near-far effect cause the performance of the conventional single user detector in DS/CDMA systems to degrade. Due to high complexity of the optimum multiuser detector, suboptimal multiuser detectors with less complexity and reasonable performance have received considerable attention. In this paper, we analyze and examine the performance of multilayer perceptron neural networks using back-propagation algorithm as multiuser detectors of CDMA signals in A WGN and multipath fading channels. Our results show significant improvement over the previous research. We compare the performance of neural network with the other detectors used in CDMA system. We also apply different neural networks and criterions such as the decision based, fuzzy decision, discriminative learning, minimum classification, and cross entropy neural nets and compare their performance. Further, we propose modified decision based network which improves the performance of decision based network