Research On Multi-source Information Fusion Task Scheduling Models And Algorithms Based On Space-based Information Port

With the development of space-based information system,the concept of a space-based network node with multi-source information fusion capabilities has been proposed.It provides users with data relay,on-board processing and other services in synchronous orbit.Multi-source information fusion technology,which collects multi-dimensional remote sensing data by a processing node,extracts and fuses the different features of various data,and can obtain more accurate comprehensive information.It is widely used in military,disaster emergency and other fields.In the space-based information port to achieve multi-source information fusion,the data can be transmitted,processed and distributed on the satellite,compared to realize mulit-source information fusion on the ground data center,space-based information port closer to the data source satellite,satellite data no longer need to be transmitted to the data center,reducing the transmission time of data center receive and distribute data.It can improve the timeliness of information acquisition,which has great significance for the space-based information system construction.Considering the high timeliness requirement of military and disaster emergency tasks,how to rationally plan tasks under the limited space-based information resources and improve the timeliness of tasks has become an important problem for the study of space-based information port,but so far there has been no such research.Based on the present situation,this paper proposes a multi-source information fusion task scheduling problem based on space-based information port: the mutil-source information fusi-on task is divided into a group of sub-tasks related to the data source satellites,including data transmission stage and data processing stage,and is sequential scheduled on antenna resources and processing resource.How to allocate the execution time of each sub-task on antenna resources and processing resource under the constraints of resources and tasks,so as to minimize the completion time of multi-source information fusion task.The establishment of task scheduling model and the algorithm of space-based information port will be the key to the task scheduling of space-based information port,and the main works are as follows:Firstly,a multi-source information fusion task scheduling model is established by analyzing the task flow,resource characteristics of the space-based information port and constraint of time window,and the optimization objective is to minimize the task completion time.Secondly,on this basis of space-based information port task scheduling model,a multi-machine circular insertion algorithm is proposed to solve the model.The simulation results show that the algorithm can reduce the task completion time by 10.8% compared with the list scheduling algorithm,and the running time is about 1/20 of the genetic algorithm.It is verified that the algorithm can meet the high timeliness of task scheduling of space-based information port and is of great significance for the multi-source information fusion in the space-based information port.Lastly,considering the limited storage resources,we study the scheduling problem of space-based information port task based on storage constraints,establish a model and propose an improved imperial competitive algorithm to solve.The simulation results show that the algorithm can get the minimum task completion time in feasible running time compared with the genetic algorithm and the imperial competitive algorithm,and verify that the storage constraint will increase the task completion time.