Paired semi-blind channel identification for wideband CDMA communications

Semi-blind channel identification methods have successfully been applied to 3G W-CDMA communication systems. However, the computational complexity is relatively high because channel coefficients are estimated in high dimensional space. In this paper, a paired semi-blind channel identification method is proposed to reduce the computational complexity dramatically while maintaining the original performance. First, an equivalent SIMO channel model is derived for the de-spread signals through employing a dual-antenna receiver. Second, a paired semi-blind channel identification scheme is obtained by applying both the cross-relation method and pilot symbols to each pair of channel coefficients with the same multipath delay across two antennas. This allows one to estimate the channel coefficients in 2-dimensional space. Finally, simulations are performed to evaluate the paired semi-blind channel identification method´s applicability to 3G W-CDMA communication systems