Phase-Reconfigured Magnitude-Selective Affine Function-Based Digital Self-Interference Cancellation System

In 5G new radio, the number of band combinations increases and bandwidths are wider to meet a demand for a high data rate. As a result, a structure of a mobile front-end has become complex. Hence, transmitter selfinterferences have significantly increased and their models have become more complicated. To efficiently approximate a highly nonlinear interference model, a phase-reconfigured magnitudeselective affine (PR-MSA) function is proposed. The PR-MSA model is a modified magnitude-selective affine model to resolve a phase transition of input signals caused by an intermodulation at interference frequencies. In addition, a digital selfinterference cancellation (DSC) system using PR-MSA model is proposed to verify a modeling capability. Experimental measurements have shown that a normalized mean square error of PR-MSA model is improved up to 1.2 dB for the highly nonlinear interference with lower complexity compared to a memory polynomial-based model. And the proposed PR-MSA function-based DSC has demonstrated an improvement of $\mathbf{0. 2 5 d B}$ in signal-to-noise at uncoded bit-error-rate.