Research On Interference Resource Scheduling Based On Heuristic Algorithm

At present,the application of UAVs in various fields is more and more extensive,and the reliability and safety problems brought by UAVs are gradually increasing.The "black flying" UAVs bring serious personal safety and privacy issues to various no-fly zones.How to efficiently use anti-UAV equipment to ensure the safety of no-fly zones,important places and facilities has become a research hotspot.Due to its flexibility and low-cost advantages,the way of UAV action gradually begins to favor coordinated action.Meanwhile,the rise of UAV swarms and the complexity of the electromagnetic environment make it necessary to schedule multiple interference resources to carry out cooperative interference,which requires to schedule one’s own interference resources according to their attribute information within a limited time.The resource scheduling problem is one of the important research points in the field of interference countermeasures.This problem is a NP combinatorial optimization problem.Due to the high computational complexity,there is no polynomial time algorithm to obtain the optimal solution.Focusing on the problems of the existing interference resource scheduling research,such as simple constraints,small confrontation scale,lack of timeliness for accurate mathematical algorithms,and failure to consider real-time dynamics,this paper conducts the following work and contributions:First,a benefit-cost multi-objective optimal interference resource scheduling model is established that constrains the minimum number of objectives to be allocated resources.To solve the problems that the heuristic algorithm converges slowly and tends to fall into local convergence,a parallel genetic algorithm with an elite set working in a hybrid model is designed.Firstly,the situation analysis is carried out according to the electromagnetic environment,and the interference effect and cost are classified and quantified.According to the scheduling requirements of UAV interference resources,the minimum number of targets to be allocated interference resources,the number of resources that can be called,and the interference correspondence are constrained.For the proposed multi-objective optimization model,this paper proposes an improved genetic algorithm for the parallel model,which improves the diversity of the population.In addition,an elite set whose aim is to reserve excellent individuals is added to the information exchange of the mixed model to reduce the probability of local convergence and speed up the convergence.The simulation results show that the proposed algorithm has a faster solution speed and higher objective function value than the non-parallel improved genetic algorithm.Secondly,in order to solve the problem of low timeliness of heuristic algorithms based on overall resources to solve large-scale resource scheduling problems,an optimal scheduling model based on potential game is proposed,and a game theory algorithm based on taboo is designed.The algorithm uses the divide--and-conquer strategy to calculate the individual optimal solution of each resource,and regard each interference resource as a participant in the game.Then,the target drone that can be interfered by the interference resource is used as the strategy of the game participant.The objective function value of the generated benefit-cost joint optimization is used as the benefit of the game participants.The potential game model is further established.It is proved that the game is a potential game and has a pure strategy Nash equilibrium solution.The simulation results show that,compared with the heuristic algorithm that takes the whole resource as the object and the game theory algorithm that does not use the potential game,the proposed potential game model and algorithm achieve better solution speed and higher the final objective function value after convergence.Finally,in view of the problem that the existing adversarial resource scheduling only considers a single layer of reconnaissance or interference,and has insufficient adaptability to the dynamic environment,a reconnaissance-jamming two-layer dynamic resource scheduling model is proposed.Accordingly,a two-layer dynamic resource scheduling algorithm with reconnaissance-jamming integrated aircraft is designed to optimize the solution of the model.Firstly,a reconnaissance resource scheduling model of benefit-cost joint optimization is established,and a solution algorithm of dynamic two-layer optimization is proposed by combining the jamming resource scheduling model and the characteristics of reconnaissance UAV,jamming UAV,and reconnaissance-jamming integrated aircraft.After the initial reconnaissance scheduling is completed,the reconnaissance and interference scheduling results can be output at the same time.The algorithm can be interrupted in real time when the information is updated.Besides,it can detect the information update after a single scheduling,so as to adapt to the dynamic scheduling scenario.Simulations show that the reconnaissance and jamming resource scheduling results of the algorithm can converge in both phases.The algorithm also has a short interrupt response time.