Reduced Complexity Subspace Based Blind Identification for DS/CDMAy

Direct-Sequence/Code-Division Multiple-Access (DS/CDMA) is emerging as a potential multiple access communication scheme for future digital wireless communications systems. Such wideband systems usually operate in a frequency selective fading channel, which introduces inter-symbol interference and thus potential performance degradation. Previously proposed subspace based blind channel identiication algorithms, which provide estimates of channel parameters for eeective equalization, suuer from high numerical complexity for systems with high spreading gain. In this paper, it is shown that, through the use of matched lter outputs, reduction in numerical complexity can be obtained. The complexity reduction is considerable when the channel length is small and the system is under-loaded. The results show that the new algorithm suuers a slight performance loss. Although the matched lter outputs do not form a set of suucient statistics for the unknown channels, the diierence between the matched lter outputs and the suucient statistics becomes negligible for large observation lengths. Performance is evaluated through simulations, the derivation of a tight approximation of the mean-squared error of the channel estimation, and through comparisons to the Cram er-Rao bound for the estimation error. Finally, it is shown that approximation of the mean-squared error can be obtained in terms of the correlation of the spreading codes and channels. This representation of the error supplies a tool for investigating the relationship between performance and spreading sequence correlations.