Digital Self-interference Cancellation Algorithm Based on The Movable Shape Filter

Abstract

In digital self-inference cancellation field, the adaptive filter is the main way to eliminate self-interference. A mass of research is completed to optimize the performance of the adaptive algorithm. Nevertheless, we have noticed there are few scholars to modify the shape of the filter to achieve a better cancellation effect. Constrained by the fixed and immovable shape, the typical adaptive filter can't full use the information behind the digital signal. Stimulated by this problem, a lot of work have been fulfilled in order to acquire a more flexible filter shape. In this paper, we propose a novel filter architecture named as the movable shape filter. Through adding the position parameters to the typical filter, finer elimination results have been reached. Moreover, the rigorous simulation results show a great progress in the mean square error.