Path Computing Scheme with Low-Latency and Low-Power in Hybrid Cloud-Fog Network for IIoT  

作　　者：Jijun Ren Peng Zhu Zhiyuan Ren 

机构地区：[1]School of Communications and Information Engineering,Xi’an University of Posts&Telecommunications,Shaanxi 710061,China [2]State Key Laboratory of Integrated Services Networks,Xidian University,Shaanxi 710071,China

出　　处：《China Communications》2023年第8期1-16,共16页中国通信（英文版）

基　　金：supported by the Shaanxi Key R&D Program Project(2021GY-100).

摘　　要：With the rapid development of the Industrial Internet of Things(IIoT),the traditional centralized cloud processing model has encountered the challenges of high communication latency and high energy consumption in handling industrial big data tasks.This paper aims to propose a low-latency and lowenergy path computing scheme for the above problems.This scheme is based on the cloud-fog network architecture.The computing resources of fog network devices in the fog computing layer are used to complete task processing step by step during the data interaction from industrial field devices to the cloud center.A collaborative scheduling strategy based on the particle diversity discrete binary particle swarm optimization(PDBPSO)algorithm is proposed to deploy manufacturing tasks to the fog computing layer reasonably.The task in the form of a directed acyclic graph(DAG)is mapped to a factory fog network in the form of an undirected graph(UG)to find the appropriate computing path for the task,significantly reducing the task processing latency under energy consumption constraints.Simulation experiments show that this scheme’s latency performance outperforms the strategy that tasks are wholly offloaded to the cloud and the strategy that tasks are entirely offloaded to the edge equipment.

关 键 词：collaborative offloading strategy cloudfog network architecture industrial internet of things path computing PDBPSO 

分 类 号：TN929.5[电子电信—通信与信息系统] TP391.44[电子电信—信息与通信工程]