Citation

Ahmad, Israr and Amphawan, Angela and Lim, Yau Kok and Ling, Mee Hong and Neo, Tse Kian (2025) RL-based centralized trust model for 5G and beyond networks. In: 4th International Conference on Computer, Information Technology and Intelligent Computing, CITIC 2024, 23 July 2024 - 25 July 2024, Virtual, Online.

Text Scopus4.pdf - Published Version Restricted to Repository staff only Download (216kB)

Abstract

Trust is a critical factor in network collaboration, influencing future actions and reactions based on experiences. While various trust schemes have been proposed to enhance traditional networks, the exploration of trust models in future networks is still in its early stages. Legitimate entities utilize reinforcement learning (RL) to enhance their trust models, while malicious entities exploit RL to maximize their harmful effects covertly. To address this, we introduce a centralized trust model designed to ensure network-wide trust in the presence of malicious entities. This model comprises a centralized controller coordinating network-level trust in a highly dynamic and heterogeneous network by detecting malicious entities. It adapts to the operating environment’s trust level, learning more or less, depending on its trustworthiness. The model’s feasibility relies on an RL-based centralized controller ensuring comprehensive trust by maintaining global network information. It leverages on features such as device-to-device communication, traffic offloading and bypassing centralized entities. Additionally, it addresses high heterogeneity and dynamicity arising from the fluctuating behaviors of entities and user requirements. Simulation results demonstrate the superiority of the proposed model over existing RL-QRT by up to 15% in parameters such as maliciousness detection, data transmission rate, and accuracy.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Reinforcement learning, Data processing, Learning and learning models
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science > QA76.75-76.765 Computer software
Divisions:	Faculty of Creative Multimedia (FCM)
Depositing User:	Nurin Syazwani Azmi
Date Deposited:	10 Dec 2025 08:02
Last Modified:	13 Dec 2025 11:49
URII:	http://shdl.mmu.edu.my/id/eprint/15046

Downloads

Edit (login required)