Privacy-Preserving Edge AI with Graph Transformer for Secure Real-Time Smart Engineering Education in Dense Urban Areas

Haider TH. Salim ALRikabi; Hussain Ali Mutar; Hind Sabeeh  Raheem; Huda Abbas  Kanber; Sanaa Aliwi Abdel Sada Habib; Ibtihal Razaq Niama ALRubeei

doi:10.3991/ijep.v16i6.62878

Authors

Haider TH. Salim ALRikabi Wasit University, Wasit, Iraq https://orcid.org/0000-0002-5201-2084
Hussain Ali Mutar Wasit University, Wasit, Iraq https://orcid.org/0009-0004-4934-416X
Hind Sabeeh Raheem University of Baghdad, Baghdad, Iraq
Huda Abbas Kanber University of Baghdad, Baghdad, Iraq https://orcid.org/0000-0002-6118-1613
Sanaa Aliwi Abdel Sada Habib University of Baghdad, Baghdad, Iraq
Ibtihal Razaq Niama ALRubeei Wasit University, Wasit, Iraq https://orcid.org/0009-0003-8958-8487

DOI:

https://doi.org/10.3991/ijep.v16i6.62878

Keywords:

smart engineering education; edge AI; graph transformer; federated learning; privacy-preserving analytics; dense urban campus

Abstract

Engineering education is becoming more advanced in rapidly expanding urban centers with smart connected classrooms, laboratories, wireless networks, indoor environment sensors, and digital learning platform technologies. This establishes three major challenges and requires real-time educational analytics, individual campus data protection, and a robust, deployable system capable of withstanding cyber risks, heavy congestion & heterogeneous edge hardware. However, the majority of existing literature on edge analytics and federated learning or graph-based models, or security monitoring , addresses them individually with limited attention to unifying real-time laurence of secure architectures supporting multi-buildings on engineering campuses. This paper proposes a privacy-preserving edge-graph transformer (PP-EGT) for secure real-time smart engineering education in dense urban environments. The model integrates local anonymization, temporal encoding, dynamic campus graph construction, graph-transformer reasoning, and federated secure aggregation. It jointly predicts a five-class zone state, a continuous Learning Readiness Index (LRI), and a cyber-physical incident probability using multimodal edge features from environmental, network, pedagogical, and security domains. A literature-grounded simulation benchmark representing 48 zones across six campus buildings was constructed over 12 weeks with 5 min sampling. Across five random seeds, PP-EGT achieved 91.89% macro-F1, 96.45% AUROC for incident detection, and 3.43 LRI MAE. Compared with the strongest baseline, ST-Transformer, PP-EGT improved macro-F1 by 1.45 points, reduced p95 latency from 49.59 ms to 44.02 ms, and decreased privacy-leakage AUC from 69.19% to 53.83%. Its novelty lies in jointly optimizing learning analytics, cyber-physical security, privacy, and edge deployment constraints.

Author Biography

Haider TH. Salim ALRikabi, Wasit University, Wasit, Iraq

Electrical Engineering Department, College of Engineering, Wasit University, Wasit, Iraq

References

[1] F. Kwateng and W. Y. Leong, "Enhancing engineering learning through digital twin and IoT applications: Student engagement and industry alignment in the Ghanaian context," Eurasia Journal of Mathematics, Science and Technology Education, vol. 22, no. 3, p. em2787, 2026, doi: https://doi.org/10.29333/ejmste/17970.

[2] M. U. Tariq, "Building the IoT-Enabled Smart Campus: Infrastructure, Automation, and Intelligent Connectivity," in Reinventing Higher Education Through Digital Transformation and Smart Campuses: IGI Global Scientific Publishing, 2026, pp. 103-130.

[3] V. Mezhuyev, Z. Li, C. B. Westphall, L. Uden, and S. S. Agaian, "Intelligent Transportation and Smart Cities: Proceedings of the 2nd International Conference (ICITSC 2025), Luoyang, China, 21-22 March 2025," ed: SAGE Publications 1 Oliver's Yard, 55 City Road, London, EC1Y 1SP.

[4] M. Tong and M. Chen, "Application of Digital Twin in Campus: Synergistic Effects of IoT, BIM, and GIS," in Intelligent Transportation and Smart Cities: Proceedings of the 2nd International Conference (ICITSC 2025), Luoyang, China, 21-22 March 2025, 2025: SAGE Publications 1 Oliver's Yard, 55 City Road, London, EC1Y 1SP, pp. 415-424, doi: https://doi.org/10.3233/ATDE2502.

[5] O. S. Asfour and O. E. Al-Mahdy, "Shaping the Future of Smart Campuses: Priorities and Insights from Saudi Arabia," Urban Science, vol. 10, no. 2, p. 34doi: 10.3390/urbansci10020034.

[6] K. Polin, T. Yigitcanlar, M. Limb, and T. Washington, "Smart Campus Performance Assessment: Framework Consolidation and Validation Through a Delphi Study," Buildings, vol. 14, no. 12, p. 4057doi: 10.3390/buildings14124057.

[7] J. Hong, Y. Hong, X. de Foy, M. Kovatsch, E. Schooler, and D. Kutscher, "RFC 9556: Internet of Things (IoT) Edge Challenges and Functions," ed: RFC Editor, 2024.

[8] H. A. Mutar, S. M. M. Najeeb, M. L. Aldabag, and H. T. Salim, "Hybrid Neural Network-Genetic Algorithm Framework for EEG and ECG Signal Classification," 2026, doi: DOI:10.53759/7669/jmc202606021.

[9] W. Holmes and F. Miao, Guidance for generative AI in education and research. Unesco Publishing, 2023.

[10] Unesco, Recommendation on the ethics of artificial intelligence. United Nations Educational, Scientific and Cultural Organization, 2022.

[11] P. Subramani, National Institute of Standards and Technology (NIST). (2023). AI risk management framework (AI RMF 1.0). U.S. Department of Commerce. 2014.

[12] P. J. Phillips et al., "Four principles of explainable artificial intelligence," 2021.

[13] A. Arif, M. Z. H. Ali, S. Z. Q. Haider, and Q. Niaz, "AI-Driven Edge Computing for IoT: Revolutionizing Phishing Detection and Mitigation," Journal of Computing & Biomedical Informatics, vol. 9, no. 01, 2025.

[14] M. Kwet and P. Prinsloo, "The ‘smart’ classroom: a new frontier in the age of the smart university," Teaching in Higher Education, vol. 25, no. 4, pp. 510-526, 2020/05/18 2020, doi: 10.1080/13562517.2020.1734922.

[15] H. A. Mutar and I. R. N. ALRubeei, "AI-Driven Power Optimization in 6G Networks: A Comprehensive Review," International Journal of Computational and Electronics Aspects in Engineering, vol. 7, no. 1, pp. 45-57, 2026, doi: 10.26706/ijceae.7.1.20260104.

[16] Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, and P. S. Yu, "A Comprehensive Survey on Graph Neural Networks," IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 1, pp. 4-24, 2021, doi: 10.1109/TNNLS.2020.2978386.

[17] P. Celestin et al., "Regulatory Compliance Challenges Under NIST AI Risk Management Framework for AI-Enabled Cybersecurity in Healthcare Critical Infrastructure," Artificial Intelligence for Engineering, 06/19 2025.

[18] R. Minerva, G. M. Lee, and N. Crespi, "Digital Twin in the IoT Context: A Survey on Technical Features, Scenarios, and Architectural Models," Proceedings of the IEEE, vol. 108, no. 10, pp. 1785-1824, 2020, doi: 10.1109/JPROC.2020.2998530.

[19] H. A. Mutar, I. R. N. ALRubeei, O. H. Yahya, N. A. Hussien, H. T. S. AlRikabi, and A. H. M. Alaidi, "Auto-PPA: An Adaptive Deep RL Agent for VLSI Physical Design Optimization," Journal of VLSI Circuits and Systems, vol. 8, no. 1, pp. 9-19, 2026, doi: https://doi.org/10.31838/JCVS/08.01.02.

[20] D. Amo-Filva, D. Fonseca, F. J. García-Peñalvo, M. A. Forment, M. J. Casany Guerrero, and G. Godoy, "Exploring the landscape of learning analytics privacy in fog and edge computing: A systematic literature review," Computers in Human Behavior, vol. 158, p. 108303, 2024/09/01/ 2024, doi: https://doi.org/10.1016/j.chb.2024.108303.

[21] D. Valdes-Ramirez, V. S. Caamal, J. I. H. Cardozo, and G. Zavala, "Edge Computing Model for Real-Time Classroom Behavior Analysis in Face-to-Face Lessons," in Distributed Computing and Artificial Intelligence, Special Sessions II, 22nd International Conference, Cham, G. Marreiros et al., Eds., 2026// 2026: Springer Nature Switzerland, pp. 495-505.

[22] H. Mutar, I. Alrubeei, I. Svyd, H. Alrikabi, and A. H. Alaidi, "AI-Based Model for Home Waste Separation Using Raspberry Pi 5 AI Kit," Physics and Chemistry of Solid State, vol. 27, pp. 44-51, 02/07 2026, doi: 10.15330/pcss.27.1.44-51.

[23] Y. Kong, R. Dong, and H. Zhang, "Classroom behavior analysis and digital teaching quality evaluation based on spatiotemporal graph neural network," Discover Artificial Intelligence, vol. 5, no. 1, p. 404, 2025/12/29 2025, doi: 10.1007/s44163-025-00623-z.

[24] J. Geng, B. Tang, B. Zhang, J. Shao, and B. Luo, "Demo: FedCampus: A Real-world Privacy-preserving Mobile Application for Smart Campus via Federated Learning & Analytics," presented at the Proceedings of the Twenty-fifth International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, Athens, Greece, 2024. [Online]. Available: https://doi.org/10.1145/3641512.3690630.

[25] S. A. Mahmud, N. Islam, Z. Islam, Z. Rahman, and S. T. Mehedi, "Privacy-Preserving Federated Learning-Based Intrusion Detection Technique for Cyber-Physical Systems," Mathematics, vol. 12, no. 20, p. 3194doi: 10.3390/math12203194.

[26] A. Aminifar, M. Shokri, and A. Aminifar, "Privacy-preserving edge federated learning for intelligent mobile-health systems," Future Generation Computer Systems, vol. 161, pp. 625-637, 2024/12/01/ 2024, doi: https://doi.org/10.1016/j.future.2024.07.035.

[27] H. Mutar, A. Abdullah, O. Abdaumran, I. Alrubeei, and H. Alrikabi, "Hybrid AI model-driven dynamic spectrum sharing for 6G wireless IoT networks," Sustainable Engineering and Innovation, vol. 8, pp. 53-72, 02/17 2026, doi: 10.37868/sei.v8i1.id722.

[28] J. Pu, S. Li, M. Guo, X. Chen, and Z. Xiong, "Graph neural network based intelligent tutoring system: A survey," Neurocomputing, vol. 610, p. 128442, 2024/12/28/ 2024, doi: https://doi.org/10.1016/j.neucom.2024.128442.

[29] A. Shehzad et al., "Graph Transformers: A Survey," IEEE Transactions on Neural Networks and Learning Systems, pp. 1-20, 2026, doi: 10.1109/TNNLS.2025.3646122.

[30] H. Wang et al., "Stgformer: Efficient spatiotemporal graph transformer for traffic forecasting," arXiv preprint arXiv:2410.00385, 2024, doi: https://doi.org/10.48550/arXiv.2410.00385.

[31] M. Ivanovska, J. Kreft, V. Štruc, and J. Perš, "Privacy-by-design AIoT vision for intelligent urban environments," Journal of Systems Architecture, vol. 169, p. 103586, 2025/12/01/ 2025, doi: https://doi.org/10.1016/j.sysarc.2025.103586.

[32] Y. Mochizuki, E. Bruillard, and A. Bryan, "The ethics of AI or techno-solutionism? UNESCO’s policy guidance on AI in education," British Journal of Sociology of Education, pp. 1-22, 2025/05/21 2025, doi: 10.1080/01425692.2025.2502808.

[33] I. R. N. ALRubeei, H. A. Mutar, H. T. ALRikabi, I. L. H. Alsammak, H. A. Kanber, and B. H. Majeed, "Design of a Hybrid AI-Driven Engineering Model for Energy-Efficient and Sustainable Educational Systems," International Journal of Engineering Pedagogy, vol. 16, no. 1, 2026, doi: 10.3991/ijep.v16i1.60437.

[34] H. Li et al., "Graph neural networks in intelligent transportation systems: Advances, applications and trends," arXiv preprint arXiv:2401.00713, 2024, doi: https://doi.org/10.48550/arXiv.2401.00713.

[35] A. Abdullah, H. Mutar, A. Hamad, I. Alrubeei, and H. Alrikabi, "Towards sustainable cropping: AI-driven precision agriculture for optimal water and pesticide use via drones and soil sensors," Heritage and Sustainable Development, vol. 8, pp. 161-176, 02/17 2026, doi: 10.37868/hsd.v8i1.1887.