低轨道卫星星座系统网络资源分配方法的研究与仿真实现

发布时间：2018-07-16 23:34

【摘要】：因为卫星通信网络有全球直接通信、信息快速传递、覆盖范围广和不受地理环境限制等特点,所以在军事领域和民用工程上都有着广泛的应用。在不同需求场景下研究卫星的资源分配,并使网络承载能力达到最大,对于设计出较高性价的卫星通信网络有重大的指导意义。本文用网络容量衡量整个网络的承载能力,结合低轨道卫星的通信环境,在设计了网络容量分析方法的基础上,研究了不同需求角度下卫星带宽和功率资源的分配方法。首先,建立了低轨道卫星节点模型、星际链路模型和传输模型。然后,把网络容量分为瞬时网络容量和平均网络容量,根据星座系统中星际链路的切换时刻,把网络周期划分成若干个时间片,通过在各个时间片内选取大量采样点的方法计算出时间片内的平均网络容量,再得出总体的网络容量。最后,从不同的需求角度设计了不同的资源分配方法,主要包括基于容量最大化的资源分配方法、基于效用函数的资源分配方法和基于比例公平性的资源分配方法。其中,基于容量最大化的资源分配方法是一种理想化的分配方式,只以容量最大化为目标,不考虑相邻卫星的实际需求和公平性；基于效用函数的资源分配方法是在最大化相邻卫星满意度的基础上进行资源分配；基于比例公平性的资源分配方法考虑了卫星覆盖区域内地面站的数目和相应星地链路存在时间。这三种资源分配方法都是非线性混合整数规划问题,分别用具有整数操作的多种群云差分进化算法和燕子群算法进行优化求解。为了验证本文设计的各种资源分配方法的可行性和有效性,本文利用卫星仿真工具包STK(Satellite Tool Kit)产生铱星系统和全球星系统的卫星轨道数据,在Eclipse平台上进行仿真实现和性能对比。仿真结果表明,尽管不同资源分配方法下得出的网络容量有些区别,但都体现出了卫星网络的承载能力会受到卫星的轨迹和实际业务的影响。
[Abstract]:Because the satellite communication network has the characteristics of global direct communication, rapid information transmission, wide coverage and not restricted by geographical environment, so it has been widely used in military field and civil engineering. It is of great significance to study the resource allocation of satellites in different demand scenarios and to maximize the network carrying capacity, which is of great significance for the design of satellite communication networks with higher price. In this paper, the network capacity is used to measure the carrying capacity of the whole network. Combined with the communication environment of low orbit satellite, the method of network capacity analysis is designed, and the allocation method of satellite bandwidth and power resources under different demand angles is studied. Firstly, the low-orbit satellite node model, interstellar link model and transmission model are established. Then, the network capacity is divided into instantaneous network capacity and average network capacity. According to the switching time of interstellar link in constellation system, the network cycle is divided into several time slices. By selecting a large number of sampling points in each time slice, the average network capacity of the time slice is calculated, and the overall network capacity is obtained. Finally, different resource allocation methods are designed from different requirements, including capacity maximization based resource allocation, utility function based resource allocation and proportional equity based resource allocation. Among them, the resource allocation method based on capacity maximization is an idealized allocation method, which only aims at capacity maximization and does not consider the actual demand and fairness of adjacent satellites. The resource allocation method based on utility function is based on maximizing the satisfaction degree of adjacent satellites. The method of resource allocation based on proportional fairness takes into account the number of earth stations in the satellite coverage area and the time of existence of the corresponding satellite-earth link. These three resource allocation methods are all nonlinear mixed integer programming problems, which are solved by multi-group cloud differential evolution algorithm with integer operation and swallow swarm optimization algorithm respectively. In order to verify the feasibility and effectiveness of the various resource allocation methods designed in this paper, the satellite orbit data of Iridium system and all-star system are generated by using Satellite tool Kit (STK), and the simulation results and performance comparison are carried out on Eclipse platform. The simulation results show that although there are some differences in network capacity under different resource allocation methods, the bearing capacity of satellite network is affected by satellite trajectory and actual service.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN927.2

【相似文献】