论文部分内容阅读
Rather than using existing self-interference cancellation methods, which essentially consist of reconstruction and subtraction, this paper proposes a novel approach, based on multiplication, to cancel selfinterference in the analog domain in full-duplex communications. This approach is called self-mixed selfinterference analog cancellation(SM-SIAC). Moreover, rather than using an individual analog cancellation circuit in existing self-interference cancellation methods, SM-SIAC can merge the analog cancellation circuit and the receiver. SM-SIAC is configured with three auto-tuning loops, consisting of one delay loop and two gain loops. SM-SIAC is further simplified with the Gaussian minimum shift keying(GMSK) self-interference signal.When these loops converge, the paper analyzes the cancellation capacity and derives a closed-form expression for the quadrature amplitude modulation self-interference signal and the GMSK self-interference signal. Simulation results illustrate the convergence of the gain loops and the cancellation capacity in the presence of engineering errors.
Rather than using existing self-interference cancellation methods, which essentially consist of reconstruction and subtraction, this paper proposes a novel approach, based on multiplication, to cancel selfinterference in the analog domain in full-duplex communications. This approach is called self-interference selfinterference analog cancellation (SM-SIAC). Moreover, rather than using an individual analog cancellation circuit in existing self-interference cancellation methods, SM-SIAC is merged with the analog cancellation circuit and the receiver. , consisting of one delay loop and two gain loops. SM-SIAC is further simplified with the Gaussian minimum shift keying (GMSK) self-interference signal. These loops converge, the paper analyzes the cancellation capacity and derives a closed-form expression for the quadrature amplitude modulation self-interference signal and the GMSK self-interference signal. Simulation results illustrate the convergenc e of the gain loops and the cancellation capacity in the presence of engineering errors.