Research On Contact Plan In Resource-Limited Satellite Networks

With the increasing demand for space communications,satellite communication technologies and related industries have developed rapidly in recent years.However,due to the periodic movement of nodes in the satellite network,the inter-satellite communication links between nodes will be frequently disconnected,and thus the network topology is time-varying.Moreover,because of the economic cost and the limitation on hardware,it is difficult for a satellite to establish multiple communication links to deliver data at the same time.Efficient scheduling links in such resource-limited networks becomes a challenge when a large amount of data needs to be transferred by the network.In light of this,Contact Plan Design(CPD)can develop a Contact Plan(CP)based on the optimization target,and efficiently schedule the links according to the CP.This Thesis focuses on the research of CPD in resource-constrained satellite networks.The main contents of the thesis can be divided into the following aspects:This thesis first classifies and introduces the satellite and satellite networks separately,and summarizes and elaborates the key technologies in the satellite network.Based on the development and research background of satellite networks,the background of this topic is introduced.According to the arrangement of chapters,the structure of this paper is sorted and introduced.The structure and characteristics of the delay-tolerant satellite network are summarized.The purpose of CPD is to efficient schedule links in a resource-constrained environment,which is similar to the routing in the Delay Tolerant Network(DTN).And thus,this thesis then classifies the routing strategies in the DTN satellite network.According to the technical background and characteristics of CPD,CPD is introduced,and the common modeling methods of CPD and state-of-the-art CPD methods are introduced.A genetical inspired Contact Plan(GICP)algorithm is proposed for resource-constrained in LEO satellite networks.Based on the characteristics of CPD,a system model is constructed.Based on the model,the CP is encoded and repaired.According to the optimization goal of minimizing the delivery time in the network,an evaluation function is established to evaluate each CP in the group.Based on the orbital characteristics of satellite networks,the optimization strategy of genetic operations was designed.The comparison between the fitness value and the transmission time in the simulation results verifies that the GICP proposed in this paper is more suitable for data transmission in resource-constrained satellite networks.A bidirectional directional particle swarm optimization(BPSO)algorithm is proposed for the acquisition and transmission of data in LEO satellite networks for remote sensing tasks.The system model is built according to the characteristics of different types of nodes and different tasks.The particles in the algorithm are coded based on the system model and the pre-stored network discrete topology.An evaluation function is formulated based on the historical state of each task and the time requirements of the task.The historical information is updated by the evaluation result and the history information stored in the algorithm.Utilizing the position of the current group and the feature of the link sparseness in the CP,the average position is calculated,and then the bits in the worst position that need to be corrected are obtained by comparing the average position and the worst position.The optimal position and the worst position are used to guide the optimization direction of each particle in the group,and an CP suitable for transferring data of remote sensing tasks is obtained iteratively.Compared with various classical intelligent algorithms,it shows that the proposed algorithm has better performance in terms of delivery time and arrival rate.