算网融合中的网络熵、非均衡度和复杂度  

Network Entropy,Disequilibrium and Complexity in Computing and Networks Convergence

作　　者：刘卓 李晖 LIU Zhuo;LI Hui(Graduate School of Chinese Aeronautical Establishment,Yangzhou Jiangsu 225006,China)

机构地区：[1]中国航空研究院研究生院,江苏扬州225006

出　　处：《佳木斯大学学报(自然科学版)》2025年第2期9-12,26,共5页Journal of Jiamusi University:Natural Science Edition

基　　金：国家自然科学基金项目(61661018)项目支持。

摘　　要：算力网络由于边缘节点算力的不足,需要将计算任务上载到上层节点完成计算,因此网络计算能力同节点存储和链路传输能力正相关。为了描述算力网络的整体效能,提出以网络状态作为参数的衡量体系,包括网络的熵、非均衡度和复杂性函数。给出了网络复杂度函数的上确界和下确界,并提出将下确界函数的包络函数作为新的下届,满足单值映射关系,下确界包络函数处处连续、可微。仿真结果证明多种网络状态下复杂度变化趋势。Due to the insufficient computing power of edge nodes,computing tasks need to be uploaded to upperlayer nodes to complete the computing.Therefore,the computing power of the network is positively correlated with the storage of nodes and transmission capability of links.In order to describe the overall efficiency of the computing power network,a measurement system based on the network state is proposed,including the entropy,disequilibrium and complexity of the networks.The upper and lower bounds of the network complexity function are given,and the envelope of the lower bounds function is proposed as the new one,which satisfies the single value mapping relationship,and the lower bounds envelope function is continuous and differentiable everywhere.The simulation results show the trend of complexity in various network states.

关 键 词：算网融合 算力网络 网络熵 非均衡度 复杂度 

分 类 号：TN915.81[电子电信—通信与信息系统]