VISUAL ANALYTICS OF SMART CITY IOT: A FEDERATED MULTI-AGENT APPROACH TO SCALABLE AND PRIVACY-PRESERVING DATA SYSTEMS
DOI:
https://doi.org/10.29121/shodhkosh.v7.i2s.2026.7267Keywords:
Federated Learning, Multi-Agent Systems, Smart City IoT, Privacy-Preserving Analytics, Edge Intelligence, Scalable Data Processing, Distributed AIAbstract [English]
The centralized data processing methods encounter the issues of crucial challenges in terms of scalability, latency, communication overhead, and privacy leakage. In spite of federated learning progress, current solutions are mostly based on single-model coordination, and they are not adaptive to the intelligence, inter-agent coordination, and resilience to changing urban conditions. This leaves a gap in scalable, privacy-focused, and autonomous decision-making processes of real-time smart city operations. In an attempt to fill this gap, this paper presents a Federated Multi-Agent Intelligence (FMAI) platform, in which a number of intelligent agents are implemented on the IoT edge nodes. The local learning, contextual reasoning, and task-specific optimization are accomplished by each agent and a federated level of aggregation can share knowledge globally without exchange of raw data. The framework combines multi-agent learning reinforcement learning, adaptable model weighting, as well as secure aggregation to manipulate non-IID information, dynamic workloads, and privacy limitations. Its key contributions are: (i) a federated multi-agent framework that can support heterogeneous IoT settings with scalability, (ii) a privacy-aware cooperation framework that has lower communication cost, and (iii) an adaptable coordination approach that enhances the system-level intelligence, where cities are dynamic. Simulated smart traffic, energy and environmental monitoring data experimental analysis shows that the prediction accuracy improves by +28.7, communication overhead is reduced by -41.3, converges much faster +34.9, and reduces latency by +22.6 when using federated over centralized and single agent baselines. The privacy leakage risk, as indicated by membership inference accuracy, decreased by -52.1. These results substantiate the notion that federated multi-agent intelligence can be used to provide efficient, secure, and scalable data processing, and that it can serve as an effective background of next-generation smart city IoT ecosystems.
References
Agarwal, P., Abdul Rahman, A., St-Charles, P.-L., Prince, S. J. D., and Ebrahimi Kahou, S. (2023). Transformers in Reinforcement Learning: A Survey. arXiv Preprint arXiv:2307.05979.
Al-Huthaifi, R., Li, T., Huang, W., Gu, J., And Li, C. (2023). Federated Learning in Smart Cities: Privacy and Security Survey. Information Sciences, 632, 833–857. https://doi.org/10.1016/j.ins.2023.03.033 DOI: https://doi.org/10.1016/j.ins.2023.03.033
AlTerkawi, L., and AlTarawneh, M. (2025). Federated Decision Transformers for Scalable Reinforcement Learning in Smart City IoT Systems. Future Internet, 17, 492. https://doi.org/10.3390/fi17110492 DOI: https://doi.org/10.3390/fi17110492
Alhasawi, Y., and Alghamdi, S. (2024). Federated Learning for Decentralized Ddos Attack Detection in IoT Networks. IEEE Access, 12, 42357–42368. https://doi.org/10.1109/ACCESS.2024.3378727 DOI: https://doi.org/10.1109/ACCESS.2024.3378727
Alshdadi, A. A., Almazroi, A. A., Ayub, N., Lytras, M. D., Alsolami, E., Alsubaei, F. S., and Alharbey, R. (2025). Federated Deep Learning for Scalable and Privacy-Preserving Distributed Denial-of-Service Attack Detection in Internet of Things Networks. Future Internet, 17, 88. https://doi.org/10.3390/fi17020088 DOI: https://doi.org/10.3390/fi17020088
Chen, L., Lu, K., Rajeswaran, A., Lee, K., Grover, A., Laskin, M., Abbeel, P., Srinivas, A., and Mordatch, I. (2021). Decision Transformer: ReinforcemenT Learning Via Sequence Modeling. Arxiv Preprint arXiv:2106.01345.
Dritsas, E., and Trigka, M. (2025). Federated Learning for IoT: A Survey of Techniques, Challenges, and Applications. Journal of Sensor and Actuator Networks, 14, 9. https://doi.org/10.3390/jsan14010009 DOI: https://doi.org/10.3390/jsan14010009
Feng, Y., Guo, Y., Hou, Y., Wu, Y., Lao, M., Yu, T., and Liu, G. (2025). A Survey of Security Threats in Federated Learning. Advances in Engineering Informatics, 11, 165. https://doi.org/10.1007/s40747-024-01664-0 DOI: https://doi.org/10.1007/s40747-024-01664-0
Heidari, A., and Jabraeil Jamali, M. A. (2023). Internet of Things Intrusion Detection Systems: A Comprehensive Review and Future Directions. Cluster Computing, 26, 3753–3780. https://doi.org/10.1007/s10586-022-03776-z DOI: https://doi.org/10.1007/s10586-022-03776-z
Kumar, M., and Kim, S. (2024). Securing the Internet of Health Things: Embedded Federated Learning-Driven Long Short-Term Memory for Cyberattack Detection. Electronics, 13, 3461. https://doi.org/10.3390/electronics13173461 DOI: https://doi.org/10.3390/electronics13173461
Meng, L., Wen, M., Yang, Y., Le, C., Li, X., Zhang, W., Wen, Y., Zhang, H., Wang, J., and Xu, B. (2022). Offline Pre-Trained Multi-Agent Decision Transformer: One Big Sequence Model Tackles All SMAC Tasks. arXiv Preprint arXiv:2112.02845. https://doi.org/10.1007/s11633-022-1383-7 DOI: https://doi.org/10.1007/s11633-022-1383-7
Nazir, A., He, J., Zhu, N., Anwar, M. S., and Pathan, M. S. (2024). Enhancing IoT Security: A Collaborative Framework Integrating Federated Learning, Dense Neural Networks, and Blockchain. Cluster Computing, 27, 8367–8392. https://doi.org/10.1007/s10586-024-04436-0 DOI: https://doi.org/10.1007/s10586-024-04436-0
Nguyen, D. C., Ding, M., Pathirana, P. N., Seneviratne, A., Li, J., and Poor, H. V. (2021). Federated Learning for Internet of Things: A Comprehensive Survey. IEEE Communications Surveys and Tutorials, 23(3), 1622–1658. https://doi.org/10.1109/COMST.2021.3075439 DOI: https://doi.org/10.1109/COMST.2021.3075439
Pandya, S., Srivastava, G., Jhaveri, R., Babu, M. R., Bhattacharya, S., Maddikunta, P. K. R., Mastorakis, S., Piran, M. J., and Gadekallu, T. R. (2023). Federated Learning for Smart Cities: A Comprehensive Survey. Sustainable Energy Technologies and Assessments, 55, 102987. https://doi.org/10.1016/j.seta.2022.102987 DOI: https://doi.org/10.1016/j.seta.2022.102987
Shahin, M., Hosseinzadeh, A., and Chen, F. F. (2025). A Two-Stage Hybrid Federated Learning Framework for Privacy-Preserving IoT Anomaly Detection and Classification. IoT, 6(3), 48. https://doi.org/10.3390/iot6030048 DOI: https://doi.org/10.3390/iot6030048
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). Attention is All You Need. In Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Long Beach, CA, USA.
Wen, M., Grudzien Kuba, J., Lin, R., Zhang, W., Wen, Y., Wang, J., and Yang, Y. (2022). Multi-Agent Reinforcement Learning is a Sequence Modeling Problem. arXiv preprint arXiv:2205.14953.
Yaacoub, J. P. A., Noura, H. N., and Salman, O. (2023). Security of Federated Learning with IoT Systems: Issues, Limitations, Challenges, and Solutions. Internet of Things–Cyber-Physical Systems, 3, 155–179. https://doi.org/10.1016/j.iotcps.2023.04.001 DOI: https://doi.org/10.1016/j.iotcps.2023.04.001
Yang, H., Huang, Y., Shi, J., and Yang, Y. (2023). A Federated Framework for Edge Computing Devices with Collaborative Fairness and Adversarial Robustness. Journal of Grid Computing, 21, 36. https://doi.org/10.1007/s10723-023-09658-x DOI: https://doi.org/10.1007/s10723-023-09658-x
Zhang, Z., Liu, M., Sun, M., Deng, R., Cheng, P., Niyato, D., and Chen, J. (2024). Vulnerability of Machine Learning Approaches Applied in IoT-Based Smart Grid: A Review. IEEE Internet of Things Journal, 11(18), 18951–18975. https://doi.org/10.1109/JIOT.2024.3349381 DOI: https://doi.org/10.1109/JIOT.2024.3349381
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Dr. Santosh H. Lavate, Sai Kiran Oruganti, Shakir Khan, Ravindra K. Moje
This work is licensed under a Creative Commons Attribution 4.0 International License.
With the licence CC-BY, authors retain the copyright, allowing anyone to download, reuse, re-print, modify, distribute, and/or copy their contribution. The work must be properly attributed to its author.
It is not necessary to ask for further permission from the author or journal board.
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
