Vector channel estimation and multiuser detection for multicarrier DS CDMA in time and frequency selective fading channels

In a multicarrier direct-sequence code-division multiple access (MC DS CDMA) system, different fading channels for different users and/or different carriers are correlated in general; thus a vector channel model is more appropriate than disjoint scalar channel models. For multiuser MC DS CDMA systems, we propose (1) a generalized vector autoregressive model which accounts for correlation between different user/carrier fading channels, (2) the use of a two-phased algorithm to obtain the proposed model's parameters, and (3) a receiver structure that consists of a generalized decorrelator followed by maximal-ratio combining (MRC) of uncorrelated carrier channel outputs of each user. The estimated fading coefficients provide the necessary quantities to MRC. The computer simulation results show that the proposed scheme has performance close to the case in which the channel is perfectly known, and outperforms separate scalar channel estimation case.