Study On Task Scheduling Of Earth-moon Relay Satellite

Lunar exploration,as the starting point of deep space exploration,has re-entered people’s vision in recent years.Countries have also paid more attention to lunar exploration.The number of probes on the moon will continue to increase,laying a foundation for future manned landings and establishing a lunar base.The probe on the back of the moon cannot directly communicate with the ground station because the moon’s rotation period is the same as the revolution period,so it must rely on the Earth-Moon relay satellite to communicate with ground.With the increase of probes,the number of tasks that require relay satellites to return to the ground station continues to increase,but the resources of Earth-Moon relay satellites are very limited and cannot fully meet the transmission needs of each task,so it is necessary to reasonably arrange the tasks.Earth-Moon relay tasks can be divided into real-time tasks and delay tolerance tasks(such as data transmission tasks).When there are no idle Earth-Moon relay satellite antenna resources,delay-tolerant task data can be buffered in a local probe,waiting for the Earth-Moon relay satellite to have idle antenna resources for transmission.However,when the amount of data waiting to be transmitted in the probe exceeds the user’s storage capacity,delay-tolerant tasks cannot be transmitted due to the limitation of the probe’s local storage resources and the Earth-Moon relay satellite antenna resources,which will result in task data loss.Therefore,how to reasonably schedule the Earth-Moon relay satellite tasks and reduce the amount of data loss under the constraints of the user storage limit is a major research problem in this thesis.In consideration of storage constraints,this thesis does the following three researches on Earth-Moon relay satellite task scheduling.Firstly,the static task scheduling problem in Earth-Moon relay satellite system was studied with the goal of minimizing the amount of data loss under the consideration of storage constraints.Aiming at this problem,a task scheduling model for earth-moon relay satellites was established,and an Earth-Moon relay satellite task scheduling algorithm based on discrete fireworks algorithm was proposed to solve this model.Simulation experiments show that the algorithm has better convergence than the genetic algorithm and can obtain better solutions.Secondly,the dynamic task scheduling problem of Earth-Moon relay satellites was studied under the constraints of storage.Based on the completion of static task scheduling,this thesis studied how to choose a suitable task preemption strategy.Aiming at this problem,this thesis established a dynamic task scheduling model for Earth-Moon relay satellites,and proposed a dynamic task preemption algorithm based on user storage.The simulation experiments show the advantages and disadvantages of several heuristic strategies.Finally,considering the existence of inter-satellite links in multiple ground-moon relay satellites,on the basis of the first point,constraints on link dynamics and bandwidth limitations were added.According to the characteristics of the task scheduling problem in Earth-Moon relay satellites in this scenario,a constraint satisfaction model was established with the goal of minimizing the amount of data loss.And we proposed an algorithm for random time slice selection and a fitness calculation function for firework algorithms.The feasibility of these two algorithms was verified by simulation experiments.