Multiuser detection and blind adaptive vector channel estimation for CDMA systems in mobile channels

A blind vector channel estimation (VCE) has been proposed and it outperforms the training-based VCE for the single-user multipath DS-CDMA system in the channel with time-varying normalized Doppler rate. In this paper, we extend it to the multiuser case which has multiple access interference (MAI). Therefore, we propose the use of MMSE multiuser detector to suppress MAI to get good initial channel estimation. Besides, following the joint vector channel estimation and decision-feedback MAP detector, we use the soft-input partial parallel interference cancellation (SIPPIC) to further reduce the remaining interference. The simulation results show that the MMSE-blind VCE-SIPPIC outperforms the MMSE-training-based VCE-SIPPIC.

[1]  Gregory E. Bottomley,et al.  Jakes fading model revisited , 1993 .

[2]  Peter J. McLane,et al.  Calculating error probabilities for DS-CDMA systems: when not to use the Gaussian approximation , 1996, Proceedings of GLOBECOM'96. 1996 IEEE Global Telecommunications Conference.

[3]  James S. Lehnert,et al.  Diversity combining for DS/SS systems with time-varying, correlated fading branches , 2003, IEEE Trans. Commun..

[4]  H. Vincent Poor,et al.  Probability of error in MMSE multiuser detection , 1997, IEEE Trans. Inf. Theory.

[5]  Hamid Aghvami,et al.  Performance evaluation of the decorrelating detector for DS-CDMA systems over multipath Rayleigh fading channels with AWGN , 1998, Ninth IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (Cat. No.98TH8361).