Abstract
In this paper, we propose an adaptive nonlinear filtering scheme based on a modified normalized least mean squares (NLMS) algorithm for cancellation of transmitter leakage in frequency-division duplexing (FDD) zero intermediate frequency (zero-IF) transceivers. The modified NLMS method is derived from the minimization of the squared Euclidean norm of the change in the weight vector between two consecutive iterations subject to a specific error constraint, where only the second-order intermodulation effect is considered. The most distinct feature of the proposed NLMS scheme is relieving the desired signal suppression issue associated with a previous related NLMS method. It achieves pretty good performance with low computational complexity. Computer simulation results show that the proposed one significantly outperforms the previous related work for transmitter leakage cancellation in FDD zero-IF transceivers.