Blind Adaptive Reduced-Rank Detectors for DS-UWB Systems Based on Joint Iterative Optimization and the Constrained Constant Modulus Criterion

摘要

A novel linear blind adaptive receiver based on joint iterative optimization(JIO) and the constrained constant modulus (CCM) design criterion is proposedfor interference suppression in direct-sequence ultra-wideband (DS-UWB)systems. The proposed blind receiver consists of two parts, a transformationmatrix that performs dimensionality reduction and a reduced-rank filter thatproduces the output. In the proposed receiver, the transformation matrix andthe reduced-rank filter are updated jointly and iteratively to minimize theconstant modulus (CM) cost function subject to a constraint. Adaptiveimplementations for the JIO receiver are developed by using the normalizedstochastic gradient (NSG) and recursive least-squares (RLS) algorithms. Inorder to obtain a low-complexity scheme, the columns of the transformationmatrix with the RLS algorithm are updated individually. Blind channelestimation algorithms for both versions (NSG and RLS) are implemented. Assumingthe perfect timing, the JIO receiver only requires the spreading code of thedesired user and the received data. Simulation results show that both versionsof the proposed JIO receivers have excellent performance in suppressing theinter-symbol interference (ISI) and multiple access interference (MAI) with alow complexity.