Biography
Enrollment Date: 2011
Graduation Date：2014
Degree：M.S.
Defense Date：2014.05.28
Advisors：Chun Zhang
Department：Institute of Microelectronics，Tsinghua University
Title of Dissertation/Thesis：Research on Interference Cancellation for Single Channel Full Duplex Communication
Abstract:
Most of the wireless communication networks are using half-duplex scheme, which can decrease the interferences between counter-propagating signals, but results in inefficient use of channel resources. Recently, full-duplex wireless communication has been proposed and attracted a great deal of attention. Full-duplex allow transmitting and receiving simultaneously, which can improve the capacity of channel greatly.
One serious problem in implementing full-duplex communication is the interference of the local transmitted signal on the signal of interest at the receiver end. In this thesis, two interference cancellation circuits are designed and implemented in RF and baseband domain respectively.
In this thesis, a single-antenna architecture is proposed for RF interference cancellation circuit. Based on software radio technologies, a full duplex testing platform is designed and realized, and a RF interference cancellation circuit is built between the antenna and the radio platform, which is used to eliminate the interference signal and improve the receiver sensitivity. Conjugate gradient method are proposed for automatically adjusting ability of the interference cancellation circuit, i.e., Self-calibration. In experiments, an average cancellation of 49dB=10MHz are obtained at the 2.4GHz frequency band.
In the baseband, the Adjustable Noise Cancellation technology (ANC) is introduced to design the interference cancellation module, which is used to eliminate the residual interference after RF cancellation. The Sign-Correlation of VSS-NLMS algorithms are first proposed for adjusting parameters of filter in the ANC module. In this thesis, the transmitter and two kinds of frame synchronization detection modules on the basis of correlation and Multi-instance Learning method are designed and realized. The whole baseband circuits are implemented in the FPGA platform.
Finally, the simulation system of full duplex wireless communication consisting of both RF and baseband interference cancellation modules are accomplished. The experimental results demonstrate that the receiver sensitivity at -100dBm meets the need of the requirements of this project.