Blind Multiuser Detection Based on Tikhonov Regularization

Abstract

In code-division multiple-access (CDMA) system, ill-conditioning of the covariance matrix of the received signal may degrade the performance of minimum mean-squared error (MMSE) detector, especially when signal-to-noise ratio (SNR) is high and small data sets are available for covariance matrix estimation. In order to mitigate the performance degradation, we propose a blind multiuser detection method based on Tikhonov regularization. Comparison between the regularization-based detector and the common subspace-based blind multiuser detector (SBMUD) is carried out. Simulations demonstrate the effectiveness of the regularization method.