Classification Studies And Applications Of Self-interference Cancellation In Full-duplex Wireless Communication System

As one of the key technologies of 5G,the Co-time Co-frequency Full-Duplex technology can transmit and receive signals at the same time and same frequency.When compared with Time Division Duplexing and Frequency Division Duplexing systems,it can obtain a double frequency spectral efficiency in best case.The Co-time Co-frequency Full-Duplex system is one of the most efficient methods to improve the capacity of the next generation system in wireless communication.However,serious loop interference resulting from its characteristic of transmiting and receiving signals at the same time and same frequency becomes a major factor which restricts its development.In this thesis,the technology of self-interference cancellation in full-duplex system is studied,including:(1)We analyze a broad range of multiple-input multiple-output(MIMO)mitigation schemes: natural isolation,time-domain cancellation and spatial suppression.Spatial suppression can be achieved by antenna selection,beam selection,null-space projection and minimum mean square error filtering.We make use of an equivalent relay to derive the self-interference power.Targeting at minimal self-interference power,we solve optimal filters for each scheme in the case of independent,separate and joint design.The performance of mitigation schemes is evaluated and compared by simulations.The results confirm that self-interference can be mitigated effectively.(2)According to frequency domain reconstruction method,we study digital self-interference suppression.We verify digital self-interference suppression on OFDM signal.The theoretical model of digital self-interference suppression method and the capability of digital self-interference suppression are analyzed.It is proved that the estimation accuracy of the self-interference channel is the key factor which affects the performance of the self-interference suppression.Then,we research a channel estimator based on compressed sensing.Our simulation results show that,compressed sensing method outperforms the conventional least square method in terms of the self-interference suppression by providing more accurate estimates of the self-interference channel.