An analysis of adaptive multistage interference cancellation for CDMA

Abstract

Formulates a simple model for an adaptive multistage interference canceller for a code division multiple access (CDMA) system. The interference canceller estimates the power of the resolvable components for each signal at each stage of cancellation. The paper presents a closed form expression for the probability of bit error for this adaptive multistage interference canceller, using a Gaussian approximation for multiple access interference (MAI). The bit error rate (BER) after any stage of interference cancellation can be computed from the signal to noise ratio, number of users and processing gain of the CDMA system. The BER expressions are extended to derive asymptotic limits on the performance of interference cancellation as the number of cancellation stages approaches infinity, demonstrating a fundamental limit on the performance improvement that can be expected from any multistage interference cancellation scheme. Furthermore, the expressions yield conditions under which interference cancellation actually hurts performance, suggesting the need for selective interference cancellation.