系统工程与电子技术 ›› 2023, Vol. 45 ›› Issue (8): 2588-2596.doi: 10.12305/j.issn.1001-506X.2023.08.34

• 通信与网络 • 上一篇    下一篇

基于节点复合特性的端-边协同网络生成算法

张冰雪1, 李希胜1,2,*, 尤佳1, 宋委任1   

  1. 1. 北京科技大学自动化学院, 北京 100083
    2. 北京市工业波谱成像工程技术研究中心, 北京 100083
  • 收稿日期:2022-08-11 出版日期:2023-07-25 发布日期:2023-08-03
  • 通讯作者: 李希胜
  • 作者简介:张冰雪 (1997—), 女, 博士研究生, 主要研究方向为物联网与边缘计算、协作网络管理
    李希胜 (1969—), 男, 教授, 博士, 主要研究方向为先进传感技术、多传感器信息融合
    尤佳 (1974—), 女, 副教授, 硕士, 主要研究方向为物联网与边缘计算、智能感知
    宋委任 (1997—), 男, 硕士研究生, 主要研究方向为Spark分布式系统的加速与优化
  • 基金资助:
    国家重点研发计划(2019YFB2101900)

End-edge collaborative network generation algorithm based on node composite characteristics

Bingxue ZHANG1, Xisheng LI1,2,*, Jia YOU1, Weiren SONG1   

  1. 1. School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, China
    2. Beijing Engineering Research Center of Industrial Spectrum Imaging, Beijing 100083, China
  • Received:2022-08-11 Online:2023-07-25 Published:2023-08-03
  • Contact: Xisheng LI

摘要:

针对智慧交通系统数据处理量大、对传输时间要求严苛的难题, 通过将协同系统定义为复杂网络模型, 构建了设备端和边缘节点协作运行的端-边协同系统。提出了基于网络节点复合特性的协作网络生成算法, 实现了多层复杂网络的构建。相比随机几何法, 提出的由节点聚类系数和介数中心性两种特性参数相结合的协作网络生成算法能够更全面地描述节点特性, 反映节点重要性, 改善现有多层复杂网络模型对节点信息描述不全的问题。最后, 使用4种复杂网络模型构建多层网络模型并验证该算法, 以节点间最短路径值作为评价指标, 证明提出的由节点复合特性生成协作网络的算法能够明显降低整个网络中节点间最短路径值, 减少传输时间, 提高传输效率。

关键词: 协同计算, 复杂网络, 网络模型, 边缘计算

Abstract:

For the difficult problem of large data processing volume and strict transmission time requirement of intelligent transportation system, an end-edge collaborative system with collaborative operation of device end and edge nodes is constructed by defining the collaborative system as a complex network model. A collaborative network generation algorithm based on network node composite characteristics is proposed to achieve multi-layer complex network's sconstruction. Compared with the random geometry method, the combination of two characteristic parameters, node clustering coefficient and the betweenness centrality, to generate the collaborative network generation algorithm can describe the node characteristics more comprehensively, reflect the node importance, and improve the problem of incomplete description of node information in the existing multilayer complex network models. Finally, use four complex network models to build multi-layer network model to verify the proposed algorithm. Taking the shortest route value between nodes as evaluation indicators, the proposed node composite characteristics can significantly reduce the shortest road value between nodes in the whole network, which could reduce transmission time and improve the transmission efficiency.

Key words: collaborative computing, complex network, network model, edge computing

中图分类号: