Low complexity quasi-optimum Decision Feedback Equalizer design for high rate ultra-wideband communication

Parameter design of Decision Feedback Equalization (DFE) in high rate DS-UWB system is involved with a challenge in the absence of practical optimized strategy, while the parameter group including the feedforward filter (FFF) length N_f, the feedback filter (FBF) length N_b, and decision delay D greatly affects both performance and complexity. In this paper, an efficient method aiming at achieving quasi-optimum performance with the low hardware cost is developed. First, the complicated equivalent statistic channel before DFE is modeled and its length is derived from the introduced energy criteria by semi-analytical approach. Second, based on the obtained N_b, and Branch-Tree-Search(BTS) theorem, we introduce a new rule SIR_DFE, to global search the optimized parameters. Numerical simulation results show that our strategy is both efficient and effective.