Research On Intelligent Anti-jamming Method For OFDM System Based On SDR

Orthogonal frequency division multiplexing(OFDM)is a multi-carrier modulation technology,which has the characteristics of fast transmission rate and strong anti fading,and is widely used in military and civil communication fields.However,with the improvement of information technology,the electromagnetic environment has become worse and worse,and the wireless communication system is facing more and more severe test.The communication jamming signals exerted by communication jamming equipments generally include single-tone jamming,multi-tone jamming and partial band jamming,etc.These interferences will increase the bit error rate of OFDM system and even make it impossible to communicate.In the face of jamming,how to realize high-speed,efficient and high-quality communication is the key problem in this thesis.Three solutions are adopted in this thesis:(1)By adjusting the modulation mode and adopting efficient modulation mode can effectively improve the communication speed of OFDM system;(2)By adjusting the transmit power,the OFDM system can achieve better communication effect with smaller transmit power;(3)By adjusting the sub-band and changing the allocation scheme of OFDM information sub-band,the jammings can be avoided flexibly and effectively to ensure high-quality communication.However,in the actual OFDM system,these three schemes can not be considered at the same time,so it is necessary to choose the three adjustment methods by an intelligent decision-making method.Reinforcement learning(RL)is one of the machine learning methods,which can maximize the return or achieve specific goals through learning.Therefore,according to different types of interference,in this thesis,an intelligent decision on modulation mode,transmit power and subband position of OFDM is made to achieve the optimal communication performance.Firstly,the theory of OFDM intelligent anti-jamming method is investigated,and an OFDM intelligent anti-jamming system model is designed based on reinforcement learning.By adding an reinforcement learning algorithm to OFDM system,the state set composed of power,subband allocation,modulation mode and the action set composed of OFDM adjustment scheme are established.Through training,the OFDM system can make an intelligent decision-making choice of power,subband position,modulation mode and other parameters in the face of different types of jammings.Then,the OFDM intelligent anti-jamming method is verified via the Simulink simulation platform.The anti-jamming performance of different OFDM modulation technologies and subband adjustment technologies are analyzed and compared.Finally,the OFDM intelligent anti-jamming method is verified based on the software defined radio(SDR)platform.An experimental verification system of OFDM intelligent anti-jamming is set up using GNU Radio software radio platform and USRP equipment,and its anti-jamming performance is verified in the actual electromagnetic environment.The results show that the proposed OFDM intelligent anti-jamming method has good anti-jamming effect.It can effectively resist partial band jamming and multi-tone jamming without significantly improving the transmitter transmission power.It can still maintain a low bit error rate when using high-speed modulation mode and low transmission power.The intelligent anti-jamming system designed in this thesis provides the basis for the research of communication anti-jamming in the actual environment,and has a certain reference value.