Research Authors & Affiliations

Abdullah Rakib Akand¹, Saiful Islam², Md. Rayanuzzaman Bishwas³, Sarminda Islam⁴, & Khandakar Rabbi Ahmed⁵

¹Department of Computer Science and Engineering (CSE), Asian University of Bangladesh, Ashulia, Dhaka, Bangladesh
²Faculty of Computing, Engineering and Science, University of South Wales, Pontypridd, CF, United Kingdom
³Department of Computer Science and Engineering (CSE), Lalon University of Science and Arts, Kushtia, Bangladesh
⁴Department of Banking and Finance, Lalon University of Science and Arts, Kushtia, Bangladesh
⁵Miyan Research Institute, International University of Business Agriculture and Technology, Dhaka, Bangladesh

Conference Information

Published in: 2026 IEEE QPAIN
Publisher: IEEE | Location: Chittagong, Bangladesh
Conference Date: 16-18 April 2026
Added to IEEE Xplore: 11 June 2026
DOI: 10.1109/QPAIN69676.2026.11546272

The advent of sixth-generation (6 G) networks is redefining mobile banking and peer-to-peer (P2P) payment systems, characterized by low latency and high connectivity, while also increasing the security threat. The classic perimeterbased security model does not fit a dynamic and decentralized environment. In this regard, a Zero-Trust Security Architecture for the mobile financial network enabled by the 6 G network will be presented, enabling behavior-aware fraud detection via the Zero-Trust Session Graph Transformer Network (ZT-SGTNet). Continuous verification is performed by enforcing the framework through user activity modeling sessions, based on the principles of zero trust. Each transaction session will be represented as a directed heterogeneous graph to capture structural, temporal, and behavioral dependencies in the data. An experimental study using a large-scale benchmark data set from the prominent PaySim platform shows that the traditional ML baselines are outperformed by the proposed method across all metrics, including accuracy (99 %), precision, recall, and F1-score. We verify the potential to guarantee next-generation 6 G mobile payment systems in an environment with zero trust principles by considering graphs at the session level alone.