Research On Digital Self-interference Cancellation In Co-time Co-frequency Full Duplex Systems

With the rapid development of modern communication technology,the demand for user speeds and services is increasing,and wireless spectrum resources are increasingly in short supply.Co-time Co-frequency Full Duplex(CCFD),as a key technology for improving the utilization of spectrum resources,is one of the important contents of the current 5th generation(5G)communication research.However,the main problem that needs to be solved to achieve this technology lies in suppression of self-interference.The current self-interference suppression adopts three methods: antenna,analog and digital.The digital method is the last-step suppression method to eliminate residual self-interference and is essential to improve the overall elimination performance of the system.This paper focuses on the study of digital self-interference cancellation,and the specific contributions are as follows:First,the paper analyzes the traditional digital self-interference cancellation method,and illustrates that the existing methods have the problems of poor real-time performance and low accuracy.Therefore,an adaptive filtering scheme based on channel estimation is proposed on the basis of the existing scheme,and combined with the experimental simulation and comparison,it is verified that the improved scheme can achieve lower elimination error than the existing scheme.Then,the paper studies the core of the proposed method,and discusses the algorithm selection,performance optimization and self-interference cancellation effect of the adaptive interference cancellation module in detail.Based on the theoretical analysis and experimental simulation,it is verified that when the correlation matrix of filter input signals has large eigenvalues,The use of the least mean square(LMS)algorithm can not balance the steady-state performance and convergence speed,while the LMS-Newton algorithm can overcome this contradiction very well.However,the convergence of algorithm is still unsatisfactory,so the improvements have been made in the original form to further increase the convergence speed of the algorithm.At the same time,the simulation results show that the LMS-Newton algorithm can achieve better self-interference cancellation effect when applied to the improved scheme.Finally,the paper considers the digital self-interference cancellation in non-ideal state.Due to the fact that there may be insufficient antenna cancellation in actual applications,higher signal power entering the receive link will result in non-linear distortion of the associated components.To this end,an estimation algorithm to suppress the nonlinear distortion is proposed.The non-linear term is reproduced in the received link signal so that the nonlinear distortion is eliminated simultaneously in the self-interference cancellation.The simulation results show that the parameter estimation obtained by the suppression algorithm is applied to adaptive interference cancellation,which improves the self-interference cancellation performance of the system effectively.