Edge-computing-based multi-objective equilibrium task unloading of Internet of Vehicles

Linjiao Wang; Chen Tang; Hua Guo

doi:10.1117/12.2685515

15 August 2023 Edge-computing-based multi-objective equilibrium task unloading of Internet of Vehicles

Linjiao Wang, Chen Tang, Hua Guo

Proceedings Volume 12719, Second International Conference on Electronic Information Technology (EIT 2023); 127194P (2023) https://doi.org/10.1117/12.2685515
Event: Second International Conference on Electronic Information Technology (EIT 2023), 2023, Wuhan, China

Abstract

The requirements of the Internet of Vehicles on the quality of service (QoS) of computing task offloading are constantly improving. The processing mode of edge computing, which deploys computing, bandwidth and storage resources near the user side, effectively improves the user experience. However, the computing capacity of edge servers is limited. Task offloading requests are overloaded and inefficient due to uneven server loads. Traditional cloud computing consumes a lot of energy and time. To solve the above problem, a KDQN based edge computing balanced task unloading mechanism is proposed: Q learning and convolutional neural network are used to obtain the optimal unloading strategy according to the returns of each step and continuous iterative updates, which reduces the convergence time of task unloading; K-means clustering is used to distribute part of the load of the overloaded server to the server with the closest European distance to achieve server load balancing. The experimental results show that compared to the existing offload mechanism, the new mechanism can reduce the average server energy consumption by 12.8%, the average service delay by 8.2%, and the average overload rate by 29.8%

Citation Download Citation

Linjiao Wang, Chen Tang, and Hua Guo "Edge-computing-based multi-objective equilibrium task unloading of Internet of Vehicles", Proc. SPIE 12719, Second International Conference on Electronic Information Technology (EIT 2023), 127194P (15 August 2023); https://doi.org/10.1117/12.2685515

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available