Diversity-assisted channel estimation and multiuser detection for downlink CDMA with long spreading codes

In downlink communication of a direct-sequence (DS) code-division multiple-access (CDMA) system, each user's short spreading codes are superimposed by base station's common long codes. This situation creates much difficulty in blind signal detection when multipath propagation occurs. However, when spatial/temporal diversity is available at the receiver, it is shown in this paper that subspace technique can be directly applied to estimate the common downlink multipath channel. Then, typical linear receivers, such as zero-forcing (ZF), minimum mean-square-error (MMSE) and RAKE receivers can be designed to detect the desired signal. Since the data covariance matrix is used but estimated from finite data samples, performance of both channel estimator and receivers gets perturbed. It is thus thoroughly and jointly analyzed by perturbation analysis. Justification of analysis and comparison of different receivers are also made through simulations.