Analysis of Cancellation Error for Successive Interference Cancellation with Imperfect Channel Estimation Project Report EE-381K: Multiuser Wireless Communications

—Successive interference cancellation (SIC) is a technique, which can increase the capacity of a DS-CDMA system. Increase in capacity has been shown to approach theoretical limits for an AWGN channel [1]. For a practical channel from the viewpoint of cellular systems, significant increase in capacity using SIC has been shown in [2] even in the presence of estimation errors up to 50%. This capacity increase in [2] is maximized if the statistics of cancellation error is known. In this paper, cancellation error for SIC for the system model proposed in [2] has been analyzed mathematically and verified through simulation for an Additive White Gaussian Channel (AWGN) an also for a fading channel. The results have shown that statistics of cancellation error in SIC has a zero mean with standard deviation and probability density function dependent on the channel statistics. The knowledge and insight gained about cancellation error statistics provide better power allocation schemes required for a typical system using SIC. The results can be used to model the estimation of fractional cancellation error for each user. It is shown in [2] that the knowledge of fractional cancellation error is important for optimum power allocation scheme in a system employing SIC. The results also suggest that better power control schemes, used in 3G systems in the uplink, can reduce the estimation error and thus provide greater capacity enhancements relative to systems not employing SIC.