Recursive multiuser detection for DS-UWB systems

Multiple access UWB communications has recently attracted a great deal of attention. In a multiple access UWB system, multiuser detection is desirable to effectively reduce multiple access interference (MAI). A suboptimal ML detector based on a recursive convex relaxation strategy is proposed for DS-UWB systems in this paper. This algorithm employs a recursive approach to deduce a sequence of reduced-size quadratic problems, in each of which only the most probable information bits are detected. Performance results are presented which show that the proposed detector provides near-optimal performance relative to the ML detector with considerably reduced computational complexity.