Title: A computational offloading algorithm for cloud-edge collaboration in smart agriculture

Authors: Feng Li; Yiyuan Li; Yuqing Pan

Addresses: School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu, China ' School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu, China ' School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu, China

Abstract: To solve the problem that the massive amount of information and real-time processing in the IoT system puts pressure on the computing resources of the whole system, the industry often adopts the computation offloading algorithm for cloud-edge collaboration. However, at this stage, conventional computation offloading algorithms frequently fail to account for dynamic conflicts between terminal tasks and offloaded tasks, and the results of offloading are inefficient. To that purpose, this paper introduces the task volume prediction model, the response time model and the power consumption model to represent the whole computing process of agricultural activities from various angles. On that basis, the DQN algorithm is used to solve the model's mixed-integer nonlinear optimisation issue. Experiments show that the algorithm proposed in this paper can increase the offloading accuracy and the speed and accuracy of cloud-edge collaborative computing.

Keywords: smart agriculture; cloud-edge collaboration; computation offloading; deep reinforcement learning.

DOI: 10.1504/IJSPM.2024.141985

International Journal of Simulation and Process Modelling, 2024 Vol.21 No.2, pp.121 - 129

Received: 22 Nov 2022
Accepted: 05 Sep 2023
Published online: 04 Oct 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article