Font Size: a A A

Research On Service Function Chain Deployment Of Satellite Networks

Posted on:2023-09-20Degree:MasterType:Thesis
Country:ChinaCandidate:T Y J SongFull Text:PDF
GTID:2558306914464404Subject:Electronic and communication engineering
Abstract/Summary:
With the development of the Space-Air-Ground Integrated Network(SAGIN),the Network Function Virtualization(NFV)and the Software Defined Network(SDN)has been applied in satellite network scenarios.Low Earth Orbit(LEO)satellite network has attracted attention for its advantages in launch cost,service delay,and power consumption.However,due to the high-speed mobility and resource limitation of LEO satellite nodes,the Service Function Chain(SFC)faces the situation of service rejection due to migration failure.Therefore,how to deploy SFC in the LEO satellite network is one of the problems in the research of SDN/NFV-enabled satellite communications networks.This thesis studies the deployment of SFC in SDN/NFV-enabled satellite communications networks.The main work is as follows:Firstly,in the LEO node movement scenario,the network topology characteristics and node movement rules are considered,and the ANS-SM-LP algorithm is proposed to maximize service profits.In this algorithm,the node with the longest visible time is selected to deploy the network function.In link deployment,the path with the highest consistency is preferentially deployed.Simulation results show that compared with the one-step migration strategy,the ANS-SM-LP algorithm can reduce the migration times by more than 60%and increase the service profit by more than about 5 times.Secondly,in order to alleviate the problem of service denial caused by resource competition in multi-SFC deployment,this thesis introduces the definition of the cost factor and proposes a node selection migration deployment strategy based on cost factors aiming at minimizing resource load rate.This strategy uses the cost factor to evaluate the cost of SFC deployment and determine the timing of SFC migration.On this basis,the TS-ANS-SMLP algorithm based on taboo search is designed.Compared with other algorithms,the TS-ANS-SM-LP algorithm inherits the advantages of the ANSSM-LP algorithm in terms of migration times in the environment of intense resource competition and reduces the network load rate below 7%on the premise of maintaining higher service profits.Finally,in view of the lack of simulation tools for SFC deployment in the scenario of satellite networks,this thesis designs a simulation platform for mobile-awareness SFC deployment based on the satellite networks and constructs the VNF management module,the SFC management module,the network management module and the scheduling management module respectively.The simulation test verifies that the platform has good performance in operability,scalability,and supporting complex network simulation.
Keywords/Search Tags:service function chaining, network function virtualization, space-air-ground integrated networks, LEO satellite
Related items