| 摘要: |
| 在空间信息网络中，各卫星间是通过星间链路（Inter Satellite Links，ISLs）相连接的，其空间网络节点的处理能力和资源存储能力受限，网络拓扑具有高动态性，通信链路存在间歇性连接.这造成空间网络节点出现高排队时延的情况，导致网络拥塞甚至丢包，空间数据传输的可靠性下降.为了高效准确地实现网络拥塞监测，本文作者分析了空间信息网络的链路稀疏性，结合其传输方式，将链路状态检测建模为压缩感知问题，并以贪婪算法求解链路延时，进而定位拥塞链路.仿真结果证明，这种链路状态检测算法可以在较少的采样数据量的情况下，以较高的精度恢复链路延时. |
| 关键词: 拥塞监测 压缩感知 贪婪算法 |
| DOI：10.3969/J.ISSN.1000-5137.2017.01.016 |
| 分类号: |
| 基金项目:国家自然科学基金（61471236，61420106008，61671295）；111计划（B07022）；上海浦江人才计划（16PJD029） |
|
| Space information network congestion monitoring based on compressed sensing |
|
Zhang Ling, Gui Lin, Gong Bo, Luo Hanwen
|
|
School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
|
| Abstract: |
| In space information network,satellites are connected through inter satellite links.High-queuing-delay,which is caused by the restriction of the spatial network node and the dynamic network topology as well as the intermittent connectivity of communication link,is often observed among spatial network node.It means that network congestion and even packet loss occurs and results in low reliability of data transmission.Aiming at realizing network congestion monitoring efficiently and accurately,this paper analyses the sparsity of links in space information network,and the link state detection is modeled as a compressed sensing problem.Based on greedy algorithm,the paper obtains the link delay and the location of congestion links.Numerical results show that the algorithm can recover the link delay with high accuracy in the case of less sample data. |
| Key words: congestion monitoring compressed sensing greedy algorithm |
