Cloud-Edge-End Collaborative Multi-Service Resource Management for IoT-Based Distribution Grid
Feng WANG,Xiangyu WEN,Lisheng LI,Yan WEN,Shidong ZHANG,Yang LIU
DOI: https://doi.org/10.1587/transfun.2024eap1011
2024-09-03
IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences
Abstract:Feng WANG,Xiangyu WEN,Lisheng LI,Yan WEN,Shidong ZHANG,Yang LIU, Vol.E107-A, No.9, pp.1542-1555 The rapid advancement of cloud-edge-end collaboration offers a feasible solution to realize low-delay and low-energy-consumption data processing for internet of things (IoT)-based smart distribution grid. The major concern of cloud-edge-end collaboration lies on resource management. However, the joint optimization of heterogeneous resources involves multiple timescales, and the optimization decisions of different timescales are intertwined. In addition, burst electromagnetic interference will affect the channel environment of the distribution grid, leading to inaccuracies in optimization decisions, which can result in negative influences such as slow convergence and strong fluctuations. Hence, we propose a cloud-edge-end collaborative multi-timescale multi-service resource management algorithm. Large-timescale device scheduling is optimized by sliding window pricing matching, which enables accurate matching estimation and effective conflict elimination. Small-timescale compression level selection and power control are jointly optimized by disturbance-robust upper confidence bound (UCB), which perceives the presence of electromagnetic interference and adjusts exploration tendency for convergence improvement. Simulation outcomes illustrate the excellent performance of the proposed algorithm. Publication Date: 2024/09/01
computer science, information systems,engineering, electrical & electronic, hardware & architecture