Code division multiple access (CDMA) techniques have been widely employed by different wireless systems with many advantages. However, the performance of these systems
is limited by interference. A number of different interference suppression techniques have been proposed, including multiuser detection, beamforming, adaptive supervised and blind algorithms, and transmit processing techniques requiring a limited feedback channel. Recently, CDMA techniques have also been combined with multicarrier and multiantenna schemes to further increase the system capacity and performance. This thesis investigates the existing algorithms and structures and proposes novel interference suppression algorithms for spread spectrum systems.
Firstly we investigate blind constrained constant modulus (CCM) stochastic gradient (SG) receivers with a low-complexity variable step-size mechanism for downlink direct
sequence CDMA (DS-CDMA) systems. This algorithm provides better performance than existing blind schemes in non-stationary scenarios. Convergence and tracking analyses of
the proposed adaptation techniques are carried out for multipath channels.
Secondly, we propose a novel space-time adaptive minimum mean squared error (MMSE) decision feedback (DF) detection scheme for DS-CDMA systems with multiple receive antennas, which employs multiple-parallel feedback branches (MPF) for interference cancellation. The proposed scheme is further combined with multistage detectors to refine estimated symbols and provide uniform performance over all users. Simulation results show that the proposed space-time MPF-DF detector outperforms existing schemes. Thirdly, we concentrate on transmit processing techniques. A novel switched interleaving algorithm is proposed to suppress interference for DS-CDMA systems, which requires the cooperation among the transmitter and the receiver, and a feedback channel sending the index of the interleaver to be used. The best interleaving pattern is chosen by the selection functions at the receiver from a codebook known to both the receiver and the transmitter and the codebook index is sent back using a limited number of bits. Symbol-based and block-based linear MMSE receivers are designed for detection.
Fourthly, the proposed switched interleaving technique is further extended to multiple antenna multicarrier code-division multiple-access (MC-CDMA) systems. A new hybrid
transmit processing technique based on switched interleaving and chip-wise precoding is proposed for the downlink case, whereas a preprocessing technique is employed for the uplink. The simulation results show that the performance of the proposed scheme outperforms the existing chip-interleaving, conventional linear precoding and adaptive spreading techniques.
At last, we present a novel multistage receivers based on multiple parallel branches successive interference cancellation (MB-SIC) for the uplink of multiple-input multipleoutput code-division multiple-access (MIMO-CDMA) systems. The proposed multistage receivers exploit a conventional ordered SIC for the first stage, followed by a grouping detection strategy and the novel MB-SIC scheme. Then, according to a selection rule, namely, the maximum likelihood (ML) or the MMSE, the MB-SIC selects the refined
estimated vector with the best performance for the desired users antenna streams.
