Parallel interference cancellation schemes based on adaptive MMSE detection for DS-CDMA systems

Abstract

The conventional parallel interference cancellation (CPIC) detector uses matched filter (MF) detectors to achieve the data estimates, which show significant performance degradation in near-far environment, thus limiting the performance of PIC. The adaptive MMSE scheme has simple structure and outperforms the MF. In this thesis, two PICs based on adaptive MMSE detection are proposed and examined in detail. One is an adaptive PIC (APIC), which uses blind adaptive MMSE detectors instead of MFs for data estimation. Through both theoretical and numerical studies, the APIC detector is shown to be immune to the near-far problem and be able to suppress inter-cell interference. In addition, to achieve further performance improvement in asynchronous channel and multipath fading channel, an adaptive decision feedback PIC (ADFPIC) scheme based on the APIC is proposed. Through numerical simulations, the ADFPIC detector is proved to have improved BER performance.