Paper
19 July 2024 Secrecy energy efficiency maximization in D2D-enabled MEC networks: a GNN-based approach
Jingying Hu, Zhifei Zhang, Yiyang Ge, Jin Mao, Zhipeng Chu, Ke Xiong, Pingyi Fan
Author Affiliations +
Proceedings Volume 13181, Third International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2024); 131817R (2024) https://doi.org/10.1117/12.3031371
Event: Third International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2024), 2024, Beijing, China
Abstract
This paper investigates the device-to-device (D2D)-enabled mobile edge computing (MEC) network, where multiple users offload their tasks to an MEC server or their paired devices for computing. In such a network, secrecy and green communication requirements are crucial due to the presence of potential eavesdroppers and the energy demands of global carbon emissions. To this end, a secrecy energy efficiency (SEE) maximization problem is formulated by jointly optimizing the computation frequency at users, the transmit power at users and the subchannel assignments, subject to the quality of service (QoS) and the available energy constraints. To solve the formulated non-convex mixed-integer nonlinear programming problem, a graph neural networks (GNNs) and genetics based resource allocation (GGRA) method is presented, where the communication system is modelled as a directed graph with relationships among users as edges. In order to enhance the ability to process users' own features and handle deep data patterns, self-loops and the ResNet blocks in convolutions and the genetics based optimization are employed. Simulations show that the proposed GGRA improves SEE performance by about 40% compared with the traditional GNN-based method with the same energy budget.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Jingying Hu, Zhifei Zhang, Yiyang Ge, Jin Mao, Zhipeng Chu, Ke Xiong, and Pingyi Fan "Secrecy energy efficiency maximization in D2D-enabled MEC networks: a GNN-based approach", Proc. SPIE 13181, Third International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2024), 131817R (19 July 2024); https://doi.org/10.1117/12.3031371
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Lab on a chip

Copper

Energy efficiency

Genetics

Power consumption

Simulations

Data modeling

Back to Top