| 摘要: |
| 为解决卫星网络中地面中继站的优化部署问题，采用Okumura-Hata模型和对数正态阴影模型模拟信号的衰落，基于随机采样的GDEM v3(ASTER Global Digital Elevation Map v3)数据构建三维覆盖模型。地面中继站部署问题被转化为最大覆盖问题(Maximum Covering Location Problem，MCLP)和集合覆盖问题(Set Covering Problem，SCP)。利用Vincenty公式计算地面中继站与终端的距离，并引入高度、障碍物等地形约束。设计了贪心、遗传和粒子群算法求解地面中继站最优部署位置。在四川阿坝、甘孜和凉山3个自治区进行部署实验，仿真结果表明，在SCP问题上，贪心算法的部署效率最高，分别只需34、58和28个地面中继站即可实现全覆盖；而在MCLP问题上，遗传算法表现出最高的覆盖率，分别通过部署20个地面中继站可实现92.64%、72.58%和98.58%的覆盖率。所提方法综合考虑了地形、信号衰减、部署难度和成本因素，展现出良好的移植性，能够高效地在不同省份实现地面中继站的优化部署。 |
| 关键词: 地面中继站 三维优化部署 贪心算法 遗传算法 粒子群算法 |
| DOI：10.20079/j.issn.1001-893x.240401002 |
|
| 基金项目:科技部重大专项（2022YFE03050004）；人工智能四川省重点实验室基金项目（2021RZJ01） |
|
| Research on Optimization Deployment of Satellite Network Ground Relay Stations |
| YANG Ninghao,LI Yitao |
| (a.School of Automation and Information Engineering;b.Artificial Intelligence Key Laboratory of Sichuan Province,Sichuan University of Science & Engineering,Yibin 644000,China) |
| Abstract: |
| To address the optimal deployment of ground relay stations in satellite networks,the Okumura-Hata model and log-normal shadowing model are used to simulate signal fading,and a three-dimensional coverage model is constructed based on randomly sampled ASTER Global Digital Elevation Map v3(〨DEM v3) data.The problem of deploying ground relay stations is transformed into the Maximum Covering Location Problem(MCLP) and the Set Covering Problem(SCP).The Vincenty formula is utilized to calculate the distance between ground relay stations and terminals,incorporating terrain constraints such as height and obstacles.Greedy,genetic,and particle swarm algorithms are designed to solve the optimal deployment locations of ground relay stations.Deployment experiments are conducted in the three autonomous regions of Aba,Ganzi,and Liangshan in Sichuan Province.Simulation results indicate that for the SCP problem,the greedy algorithm achieves the highest deployment efficiency,requiring only 34,58,and 28 ground relay stations respectively for full coverage.However,for the MCLP problem,the genetic algorithm shows the highest coverage rates,achieving 92.64%,72.58%,and 98.58% coverage with 20 ground relay stations in each region,respectively.The proposed method comprehensively considers factors such as terrain,signal attenuation,deployment difficulty,and cost,demonstrating good portability and the ability to efficiently optimize the deployment of ground relay stations in different provinces. |
| Key words: ground relay stations three-dimensional optimization deployment greedy algorithm genetic algorithm particle swarm optimization algorithm |
《电讯技术》编辑部 