Research On Measurement And Control Resource Scheduling Prediction And Service Function Chain Optimization Mapping

With the evolution and development of network technology,both satellite communication systems and terrestrial networks are faced with the problem of limited resources,increased demand for user services,and diverse types of requirements.How to quickly and efficiently solve the problem of scheduling,allocating and deploying resources is particularly important.A good scheduling solution can maximize the use of resources to meet user needs and achieve goals economically and efficiently.Aiming at the resource scheduling problem,multi-satellite measurement and control resource scheduling is one of the main problems facing satellite networks.The traditional multi-resource joint scheduling algorithm has shortcomings such as long solution time,low efficiency,high computational cost and simple system description.The Deep Neural Network(DNN)algorithm provides a new way to solve these problems,but due to the strong correlation and constraint between data,it is difficult to simply apply the measurement and control scheduling problem to the DNN algorithm.In this paper,by discretizing the data,multiple constraints and related attributes are converted into different mark bits,and part of the data is reflected in the form of binary code to reflect the constraint relationship between the data.Therefore,the DNN algorithm can be used to solve the problem of measurement and control resource scheduling prediction.Thereby improving the efficiency of measurement and control resource utilization and automation.Simulation results verify the effectiveness of the proposed model.Regarding the problem of resource mapping,as the types of network business requirements continue to increase,network virtualization technologies have emerged at the historic moment.A group of virtual network functions interconnected by a virtual link is called a Service Function Chain(SFC).How to map SFCs in physical network has become another important research hotspot.When the SFC requests arrive,due to insufficient link resources around some functional nodes,these nodes may not be able to map and become resource fragments.Considering the relationship between delay cost,resource fragmentation,link resource utilization,and link utilization threshold,this paper designs two heuristic algorithms for offline and online scenarios.Through comparative experiments,it is found that the two heuristic algorithms have great advantages in terms of load balancing,SFC deployment success rate,and node remaining resource rate.