Steady-state analysis of the adaptive successive interference canceler for DS/CDMA signals

The adaptive successive interference canceler (ASIC) is a multistage receiver that sequentially detects and removes cochannel users from the received signal impinging on a single antenna element. Each stage of the ASIC consists of a conventional matched filter (MF) detector and an adaptive interference canceler (AIC) that employs the least-mean-square (LMS) algorithm to recursively estimate the received amplitude of the detected signal. In this paper, we investigate the performance of the ASIC using a Wiener model of convergence for the LMS algorithm, deriving expressions for the asymptotic mean and variance of the amplitude estimate and the steady-state bit error rate (BER). The analyses and computer simulations demonstrate that the performance of the ASIC exceeds that of the conventional SIC (CSIC), which utilizes the MF output as the received amplitude estimate.