Adaptive Asynchronous CCI Cancellation: Selection of the Regularization Parameter for Regularized Semi-Blind Technique

Abstract

A asynchronous interference cancellation problem is addressed when training and working data intervals are available containing the desired signal and arbitrary overlapping interference. A regularized non-iterative semi-blind estimation of the antenna array coefficients is considered, which employs the autocorrelation matrix estimation as a weighted sum of the autocorrelation matrices estimated over the training and working data intervals. On-line selection of the regularization parameter is proposed. A simple switched selection algorithm is developed based on the constant modulus criterion. It is shown by means of simulations in TDMA and OFDM environments that the adaptive regularization parameter allows significant performance improvement compared to any fixed regularization parameter in a wide range of the asynchronous interference scenario parameters