Research On Interference Cancellation Method In OFDM System

Orthogonal Frequency Division Multiplexing(OFDM)technology has been widely adopted in various wireless communication systems because of its high spectrum utilization and resistance to multipath attenuation.However,in practical communication scenario,the OFDM system is susceptible to Narrowband Interference(NBI)and the system performance will be degraded.Deep Learning(DL)uses a multi-layer neural network to learn and fit the system model,which can effectively reduce the impact of narrow-band interference.This paper studies the narrowband interference cancellation problem in OFDM systems based on Convolutional Neural Network(CNN).The main work includes the following parts:Firstly,for the scenario of anti-interference in the receiver,this paper proposes a CNN interference estimator.The interference estimator extracts interference characteristics from the processed received signals and estimates time domain interference signals.This paper demonstrates the need to estimate the interference signal from the time domain,and discusses the network design and network training of the interference estimator.In this paper,the interference estimator is introduced into the interference cancellation system in different channel environments to estimate and eliminate narrowband interference.The simulation results show that the scheme can accurately estimate the interference in different environments and improve the system performance.Secondly,for the scenario where the transceiver can cooperate and mitigate the interference,this paper focuses on the OFDM interference suppression system based on the combination of transmitter and receiver.Firstly,a CNN interference parameter detector is proposed to estimate the interference signal parameters by subcarrier.Secondly,the precoding scheme is introduced into OFDM system and can be verified to effectively suppress the interference when the interference parameters are unknown at the transmitter.Finally,an OFDM narrowband interference suppression system is proposed.The receiver feeds the estimated interference parameters to the transmitter.And the transmitter will precod the signals based on the result of interference parameters detection to effectively improve the anti-interference ability and reduce the system overhead.The simulation results show that the joint processing system can effectively estimate the interference parameters,and the precoding method can counter the influence of the interference signal when the narrowband interference parameters are unchanged.