Research On Routing And Wavelength Assignment Algorithm Of Double-layer Satellite Optical

In recent years,with the rapid development of mobile Internet,5G technology,the requirements of transmission rate,security and storage capacity of satellite are increasing.Space-based satellite network has become a new trend in current communication research.GEO satellite has a large coverage area and a stronger ability to process information,so it can plan services by integrating global information.Although LEO satellite has a weak load capacity,it can realize global communication by designing multiple satellites to form a network.With the development of intersatellite link technology,multi-layer satellite network with more flexible networking and multi-functional fusion has become a new research direction.At present,most of the link transmission in satellite communication system adopts microwave scheme with mature technology.However,with the advance of the construction of the integrated information network,the microwave communication is limited by bandwidth and speed.Using laser links to transmit services ensures the security and timeliness of service transmission,eliminates the need to divide frequency band resources,and minimizes interference during transmission.Therefore,laser satellite network has become a reliable solution to break through the bottleneck of microwave satellite network.In satellite optical networks,routing and bandwidth assignment is a very important question,which relates to whether transmission can be carried out with high success rate,high accuracy and low delay.Reasonable RWA strategy will give full play to the advantages of multi-layer satellite optical network and improve user experience.On the basis of studying the networking theory of multi-layer satellite network,this thesis focuses on the RWA algorithm of GEO/LEO doublelayer satellite network,including the link length and hop count balancing RWA algorithm based on ant colony optimization and the service classification adaptive RWA algorithm based on genetic algorithm.It solves the problems of transmission delay and imbalanced number of passing satellites in double-layer satellite optical network.The main research work is as follows:(1)Strategies for establishing and dismantling intersatellite links in GEO/LEO double-layer satellite networkIn the double-layer satellite network,the operation cycle of GEO satellite and LEO satellite is different,and links between layers switch frequently.In order to solve the problem of reliable and stable communication of GEO/LEO satellite network under the condition of heterogeneous topology and dynamic link change,a GEO/LEO satellite network structure was designed,and an inter-layer link establishment strategy of two-layer satellite network was proposed.In this strategy,LEO satellites with strong transmission capacity are selected as cluster-head satellites,and the location of cluster-head satellites is determined according to the selection mode of nine grids.In terms of inter-layer communication,the link distance between GEO satellite and LEO cluster-head satellite is used as an indicator to establish inter-satellite links.By adding standby links,GEO satellite can be connected with each LEO cluster-head satellite.The coverage performance of GEO/LEO double-layer satellite network and the link length index based on link distance are simulated.The results show that the GEO/LEO double-layer satellite network structure can cover the whole world all weather,and at least four satellites can cover the whole world at the same time,which improves the robustness of communication.Compared with the method of establishing links by running region,the proposed method reduces the average length of links by 8.8%.(2)Link length and hop count balancing RWA algorithm based on ant colony optimizationMost of the current RWA algorithms applied to satellite optical networks are aimed at single-layer LEO satellite networks,which are not applicable to GEO/LEO satellite networks.In this thesis,a link length and hop count balancing RWA algorithm based on ant colony optimization(ACO-LHB RWA)is proposed for double-layer networks.The pheromone updating rule of the ant colony algorithm is determined by the satellite hop number of the selected path,and the pheromone on each path is updated by using the global information.To solve the problem of higher link switching frequency in two-layer networks,two indicators of link length and link persistence are added into the heuristic function to guide the choice of the path.The jump probability adaptive selection scheme based on cross-layer link length is added,which can adjust the jump probability according to current business requirements and next hop node types,so as to solve the problem of unbalanced node selection caused by large inter-satellite link distance difference,and rationally use GEO satellite and LEO satellite.The ACO-LHB algorithm is simulated,and the results show that the ACO-LHB algorithm is better than the traditional ACO algorithm in terms of communication performance,and can effectively balance the requirements of delay and satellite hop count in the two-layer network,and make full use of the advantages of the two-layer network for business transmission..(3)Service classification adaptive RWA algorithm based on genetic algorithmIn order to solve the problem of the diversification of the current network business requirements and using the structure of GEO/LEO network reasonably.,a service classification adaptive RWA algorithm based on genetic algorithm(GA-SCA RWA)is proposed.When genetic algorithm is applied to dynamic resource allocation,it adopts variable length coding and takes the complete optical path to be passed by the current service as an individual to generate population.According to the demand of business transmission,all businesses are divided into three levels of high,medium and low importance.The adaptive factor λ of business classification is added to the adaptability function to make adaptive adjustment according to the change of business type.The elite retention strategy is added in the selection process.On the basis of the searching range of the path,the link length can be shortened as soon as possible by using multi-point crossing and selecting the mutation range and algorithm according to the hop number of the current path.The simulation shows that compared with other traditional heuristic algorithms,the proposed algorithm can reasonably allocate resources to different priority services and improve the transmission success rate.