Research And Validation On Key Technologies Of Self-interference Cancellation In Co-time Co-frequency Full-duplex Systems

The Co-time Co-frequency Full-Duplex system can transmit and receive signal at the same time and same frequency. When compared with Frequency Division Duplexing and Time 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 method to improve next generation system capacity in both wire and wireless communication.While, in full-duplex node, the signal being transmitted interferes the estimation of the signal of interest from far-end. How to reduce such self-interference, and what challenges are we facing? This dissertation will focus on these questions.Firstly, the radio frequency(RF) self-interference cancellation(SIC) ability is studied. In practical constraints, we analyze the influence of self-interference power, bandwidth, amplitude adjustment error and phase adjustment error on self-interference cancellation ability. We work out and test the related closed form expression. Following experiments illustrate that, with a 20 MHz interference bandwidth, a 10 d Bm self-interference signal power, a 0.005 d B amplitude adjustment error and a 005 d B phase adjustment error, the RF self-interference cancellation ability can achieve 51.2d B.Secondly, the digital self-interference cancellation is discussed. According to frequency domain reconstruction method, given the multipath Rayleigh self-interference fading channel conditions, the closed form expression, which shows the influence of self-interference power and self-interference of channel estimation error on the digital self-interference cancellation ability, is derived. In our experiment, in which the bandwidth is 20 MHz, and interference to noise ratio of self-interference signal is 40 d B, digital self-interference cancellation capacity can reach 31.2d B.Thirdly, to solve the problem: the self-interference channel estimation error reduce the digital self-interference cancellation ability, this dissertation proposed an estimation method of variable window size discrete Fourier transform algorithm, which improves the accuracy of self-interference channel estimation when interference to noise ratio is low. Numerical analysis and simulation results demonstrate that, in the Rice self-interference channel conditions, with a 15 d B interference to noise ratio, this method can bring about 2.2d B self-interference cancellation capacity rise when compared with the least square channel estimation method.Finally, according to the LTE application scenario, we combine the radio frequency self-interference cancellation with digital self-interference cancellation, and work out the single-input-single-output LTE Co-time Co-frequency Full-Duplex prototype system. The demand for self-interference ability in this system is analyzed, and then the practical self-interference cancellation method is designed. The prototype test results show that, with a-10 d Bm self-interference power, combined RF and digital self-interference cancellation capacity can reach 70 d B, and the LTE Co-time Co-frequency Full-Duplex bidirectional transmission rate doubles.This dissertation provides supports in both theory and practice for self-interference cancellation techniques of Co-time Co-frequency Full-Duplex system. Currently, part results have been verified. The finding can be applied to relay system, military wireless communication system, wireless local area networks and next-generation wireless and wired communication system.