Research On Task Allocation Model And Algorithm Of Jamming Resources In Radar Countermeasure

The existence of enemy radar in the modern battlefield has posed a serious threat to our combat equipment.How to use our limited jamming resources to allocate tasks,effectively jamming enemy radar,and maximize the overall combat effectiveness of our side is one of the difficult problems to be solved in the battlefield.In thesis,the task allocation problem of interfering resources in modern battlefield is studied.The main research contents are as follows:(1)A static task allocation model of jamming resources considering multiple factors is established.In order to solve the problem of static task allocation of jamming resources,firstly,the comprehensive threat degree of radar and its anti-jamming ability in time,space,frequency,energy,polarization and waveform are determined by fuzzy comprehensive quantification.Then,considering the jamming time,jamming power,jamming frequency,jamming style and other factors,the jamming effects of the jamming resources are determined by fuzzy synthesis.Among them,the weight coefficients of related indicators are determined by combining subjective and objective methods.Finally,a mathematical model of static task allocation of jamming resources considering jamming cost is established.(2)A hyper heuristic algorithm based on Q learning is proposed to solve the static task assignment problem of jamming resources.On the basis of establishing the mathematical model of static task allocation for jamming resources,considering the shortcomings of traditional heuristic algorithms such as poor generality and easy to fall into local optimality,a hyper heuristic algorithm based on Q learning was proposed to solve the static task allocation model for jamming resources,including the determination of upper selection strategy,the selection of solution acceptance criteria and the design of lower heuristic operator.Finally,by comparing with genetic algorithm(GA)and firefly algorithm(FA),it is verified that the proposed algorithm can improve the traditional heuristic algorithm which is easy to fall into the local optimal to some extent,and has good robustness and universality.(3)A dynamic task allocation strategy based on reinforcement learning is proposed.In view of the disadvantages of traditional battlefield dynamic task allocation,which only considers the current jamming benefits and the low efficiency of manual decision making,thesis proposes a dynamic task allocation strategy of jamming resources based on reinforcement learning.This method modeled the dynamic task allocation problem as a Markov decision process(MDP),mined the law contained in historical data through reinforcement learning,made the matching strategy have a certain foresight,provided a scientific basis for dynamic task allocation,and took into account multiple common battlefield factors such as task yield and target response rate.Finally,the proposed strategy is compared with the common strategies based on reward greed and distance greed in three aspects: target response rate,total mission return and total moving distance,and it is verified that the proposed strategy has certain advantages when considering the long-term battlefield return.This thesis explores the task allocation of jamming resources based on artificial intelligence algorithm and provides technical reference for task allocation of jamming resources in intelligent radar countermeasures in the future.