Abstract

This paper investigates learning and achievable bit error rate (BER) performance of ultra-wideband (UWB) systems that use intelligent multiuser detector (MUD) when communicating over UWB channels that experience both multiuser interference (MUI) and intersymbol interference (ISI), in addition to multipath fading. Multiple access interference (MAI) degrades performance of conventional single user detector in UWB systems. Due to high complexity of the optimum multiuser detector, suboptimal multiuser detectors with less complexity and reasonable performance have drawn considerable attention. By taking advantage of heuristic values and collective intelligence of tabu search with Hopfield neural networks (TAHNN), the proposed detector offers almost the same BER performance as a full-search-based optimum multiuser detector does, while greatly reducing computational complexity. To evaluate performance and robustness of our proposed TAHNN based MUD, we experiment with a number of test problems. Computational results show that our proposed TAHNN in almost all cases outperforms other foregoing heuristics applied to this paper.

Key words: Ultra-wideband, Tabu search method, Hopfield neural networks, multiuser detection.