Task offloading for edge computing in industrial Internet with joint data compression and security protection

With the increase of intelligent devices in the industrial Internet, the computing tasks of these devices are growing exponentially. However, due to the centralized deployment and long backhaul characteristics of cloud computing, it is difficult to meet the requirements of high real-time and high se...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of supercomputing 2023-03, Vol.79 (4), p.4291-4317
Hauptverfasser: Wang, Zhongmin, Ding, Yurong, Jin, Xiaomin, Chen, Yanping, Gao, Cong
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:With the increase of intelligent devices in the industrial Internet, the computing tasks of these devices are growing exponentially. However, due to the centralized deployment and long backhaul characteristics of cloud computing, it is difficult to meet the requirements of high real-time and high security industrial tasks. Edge computing offloads tasks to the edge side to effectively reduce latency and protect data security. In this paper, we establish an optimization model of task offloading with joint data compression and security protection. In our model, in order to solve the load problem of super-large tasks on link bandwidth in the industrial Internet, a data compression model is established by formulating the computing load of compression and decompression as a nonlinear function of the compression ratio. The model can determine the optimal compression ratio and reduce the transmission latency of the task. In addition, we establish a security protection model by setting different security levels for each task. Based on this model, tasks are offloaded to different locations to improve data security and meet the computing requirements of different tasks. To solve the task offloading strategy, we design an offloading algorithm based on the improved simulated annealing particle swarm algorithm (ISA-PSO). The simulation results show that the established offloading model has remarkable effect in data security protection, latency, and cost reductions, and the objective value is reduced by 17.41% after adding the compression model. Compared with the existing edge computing offloading algorithms, ISA-PSO has better convergence level and offloading effect, which can reduce the weighted cost by up to 27%.
ISSN:0920-8542
1573-0484
DOI:10.1007/s11227-022-04821-9