Research On Routing Algorithm Of LEO Satellite Network Based On Inter-satellite Link

With the increasing demand for communication services,Low Earth Orbit Satellite Networks(LEO-SN)have become a current research hotspot due to its features of large coverage,large capacity,low delay and low power consumption.As the key technology of LEO-SN,the inter-satellite routing technology plays an important role in improving the real-time and reliability of spatial information data transmission.However,in LEO-SN based on Inter-Satellite Link(ISL),there are not only problems of dynamic network topology changes and frequent ISL handovers,but also the problems of unbalanced load distribution on the satellite,limited on-board processing capacity,non-upgrade node hardware and diversified service types.This series of problems make the design of an effective and reliable routing algorithm become a technical challenge for LEO-SN based on ISL.This thesis focuses on in-depth research on the routing algorithm in LEO-SN,and the main contents are summarized as follows:Fristly,aiming at the problem of link and node congestion in LEO-SN,a Multi-Service Routing with Guaranteed Load Balancing(MSR-GLB)algorithm is proposed,which comprehensively considers the link's Quality of Service(Qo S)and the uneven distribution of ground users.In the MSR-GLB algorithm,the Geographic Location Information Factor(GLIF)is defined to balance the network load by scheduling ISL with a lower load.The optimization objective function is constructed by delay,remaining bandwidth,packet loss rate,and GLIF in order to characterize the inter-satellite routing problem caused by multi-service and load balancing.The information state transition rule is defined based on link Qo S and GLIF to select the best next hop satellite node,and the pheromone update rule is defined based on positive and negative feedback mechanism to accelerate the convergence of MSR-GLB algorithm.The simulation results show that the MSR-GLB algorithm can better balance the network load while meeting the Qo S requirements of different services.Secondly,aiming at the problem of limited battery lifetime and non-upgrade hardware in LEO-SN,a Location-aware Routing for Extending Network Lifetime(LRENL)algorithm is proposed.In the LRENL algorithm,the eclipse period and output power of satellite nodes are defined based on the satellite's position and attitude,and the remaining battery capacity is defined based on the energy consumption and harvesting of satellite node.A Path Multi-Attribute Decision Matrix(PMADM)is constructed by delay,eclipse time,output power,and remaining battery capacity.The Preference Ranking Organization Method for Enrichment Evaluation(PROMETHEE)is used for PMADM to obtain the best path of extending network lifetime.The simulation results show that the LRENL algorithm can better extend the network lifetime and balance the traffic distribution.