Research On Communication Anti-jamming Strategy Based On Intelligent Decision

As the electromagnetic environment in which traditional wireless communication systems work is becoming more and more complex,traditional anti-jamming technologies such as frequency hopping have certain limitations in the face of intelligent interference.It is imperative to study how to further improve the anti-jamming capability of wireless communication systems.The swarm intelligence algorithm has good adaptability and optimization effect,and is widely used in channel allocation,power allocation and other aspects.The reinforcement learning algorithm exhibits exceptional real-time decision-making capabilities when confronted with the scene of intelligent interference.Therefore,this paper focuses on communication anti-jamming strategies.Based on the idea of swarm intelligence algorithm and deep reinforcement learning,through modeling and analysis of communication systems,communication parameters and communication strategies are optimized.The principal accomplishment of this thesis is outlined as follows:(1)In this thesis,an improved cuckoo search algorithm anti-jamming decision-making method is proposed for communication parameter decision-making.The algorithm improves the convergence performance and decision-making effect of the cuckoo search algorithm by introducing the elite reverse learning strategy,the adaptive Levy flight mechanism and the beetle whisker operator.And carry out system modeling on the communication anti-jamming system,establish the transmission and reception model and interference model,propose the objective function of minimizing the transmission power and the minimum bit error rate,and optimize the communication parameters such as transmission channel,transmission power and modulation mode Adjustment to achieve antijamming decision-making.The simulation results indicate that the algorithm can effectively demonstrate the anti-jamming effect in communication anti-jamming problems.Compared with the traditional cuckoo search algorithm,it further improves the convergence speed and optimization ability.(2)Considering that in the face of communication scenarios of intelligent interference and real-time game confrontation,it is necessary to confront the interferer online in a more real-time manner.Therefore,this paper proposes an anti-interference decision-making method based on the optimized DQN algorithm.Firstly,the communication anti-jamming system is modeled,and the sending and receiving model,jamming model and time slot model are established.Then the corresponding Markov decision process and anti-jamming decision-making steps are generated for the above model.Furthermore,an optimized DQN algorithm is proposed.By introducing the greedy action selection of nonlinear changes,the exploration requirements of different iteration periods of the algorithm are balanced,and the computational workload for the Q network and channel switching has been decreased.Based on the simulation results,the algorithm demonstrates the ability to adjust to communication scenarios characterized by online confrontation,exhibiting superior anti-jamming effectiveness and convergence performance.