Iterative multiuser detection and decoding for coded CDMA in frequency-selective fading channels

Abstract

A receiver structure, which his linear computational complexity with the number of users, is proposed for decoding multiuser information data in a convolutionally coded asynchronous DS-CDMA system in multipath fading channels. The proposed receiver architecture consists of a multiuser likelihood calculator followed by a bank of soft-input soft-output (SISO) channel decoders. Information is fed back from SISO channel decoders to multiuser likelihood calculator, and the processing proceeds in an iterative fashion analogous to the decoding of turbo codes. A simplification to the above receiver structure is given also. Simulation results demonstrate that, for both receiver structures, at high signal-to-noise ratios (SNR), both multiple-access interference (MAI) and intersymbol interference (ISI) are efficiently suppressed, and single-user performance is approached. Furthermore, the proposed iterative receivers are near-far resistant.