Partial Parallel Interference Cancellation Techniques for MC-CDMA Wireless Communications

• Main Authors: Chih-Chiang Wu, 吳志強
• Other Authors: Chin-Liang Wang
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/90998399335638699404

碩士 === 國立清華大學 === 通訊工程研究所 === 91 === Multicarrier code division multiple access (MC-CDMA) is a scheme that combines CDMA and orthogonal frequency division multiplexing (OFDM) techniques. This scheme has been treated as a candidate for fourth-generation wireless mobile communication systems. One major problem associated with MC-CDMA is that its capacity is greatly limited by multiple access interference (MAI). To overcome this problem, there have been many multiuser detection approaches reported in the literature. Among them, parallel interference cancellation (PIC) and its improved techniques, such as partial parallel interference cancellation (PPIC), are of great interest due to their simpler receiver structures and much less processing latency as compared to other schemes. For a PPIC scheme, the MAI estimates are cancelled partially at each interference cancellation (IC) stage, where the cancellation ratio is dependent on the fidelity of tentative decisions in forming the MAI estimates. This thesis presents a new low-complexity approach to realizing the main PPIC operations that firstly cancels the MAI estimates completely from the received signal at each IC stage and then adds a compensated term to the resulting signal, where K multiplications of the original scheme are replaced with K additions for each IC stage. Based on this approach, two equivalent PPIC-based multiuser detectors of lower complexity are presented, one for the conventional PPIC scheme and the other for the turbo PPIC (TPPIC) scheme proposed recently. Also, a new bit estimator is derived for PPIC-based multiuser detectors to reduce the variances of the final signal estimates, where the complexity grows linearly with the total number of IC stages. The proposed bit estimator linearly combines soft information of each stage using weighting factors derived from those for PPIC. Computer simulation results show that it can enhance performance of various PPIC-based receivers, especially for the TPPIC detector with few IC stages. To further improve the system performance, we employ the powerful space-time processing techniques by placing an antenna array at the base station. To jointly perform antenna diversity reception and multiuser detection, a diversity-refined multiuser detector called the enhanced PPIC (EPPIC) scheme is proposed. The proposed diversity-refined mechanism is based on the use of the maximal ratio combining (MRC) technique to enhance the input soft information at each IC stage and thus to compensate for the lost diversity gain due to possible erroneous MAI estimates. Computer simulation results demonstrate that the proposed EPPIC scheme outperforms other similar approaches, especially in extremely high system load cases and in multirate environments (where a large number of interferers or effective interferers exist).