📄 Sciences Methods and Technologies
International Journal (SciMeTech)

Volume 2 · Issue 1 · 2026
ISSN: 3085-5284
Enhanced Federated Learning Techniques in Privacy-Preserving Data Analysis Over Transit Networks
Younes Ennajjar, Wissam AABASS, Anas Abou El Kalam
Pages 75–90 · Cadi Ayyad University, UCA, The National School of Applied Sciences of Marrakech, LaRTID Laboratory, Marrakech, Morocco
⬇ PDF
Abstract
The popularity of smart transit systems is an opportunity to acquire the vast amount of system operation data unprecedented in history, which leads to a great practicability of improving urban mobility, resource utilization and passenger service, etc. But using the data for advanced analytics and ML algorithms poses substantial privacy issues, especially when data is across multiple agencies or entities. This paper discusses federated learning (FL) as an emerging paradigm for privacy-preserving data analysis in transit systems. We discuss some of the new emerging FL techniques such as model aggregation, secure multiparty computation, differential privacy methods and explain how they will affect different smart mobility use cases. However, there are still many challenges to be addressed, including in scalability over large scale urban networks, in the handling of heterogeneity of the nodes, in the measure implemented to mitigate adversarial attacks, and in enabling cross-agency data sharing within strict regulatory environments such as GDPR. This review summarizes existing works, presents current challenges and outlines future research paths for progress towards a resilient, privacy-preserving and scalable FL solutions in the changing dynamics of urban transit.
Keywords: Federated Learning, Privacy-Preserving, Data Analysis, Intelligent Transportation Systems, Differential Privacy, Secure Aggregation

References

  1. McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. (2017). Federated learning of deep networks using model averaging. 20th International Conference on Artificial Intelligence and Statistics (AISTATS).
  2. Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patrick, J., Talwar, K., Trevithick, L., & Vassilvitskii, S. (2017). Securing the aggregation in federated learning. In: CCS 17: 2017 ACM SIGSAC Conference on Computer and Communications Security.
  3. Dwork, C., Roth, A., et al. (2014). Theoretical foundations for differential privacy. Foundations and Trends in Theoretical Computer Science, 9(3-4), 211-407.
  4. Gentry, C. (2009). Fully Homomorphic Encryption from Ideal Lattices. In Proceedings of the 41st Annual ACM Symposium on Theory of Computing.
  5. Zhang, S., Li, J., Shi, L., Ding, M., Nguyen, D. C., & Dobre, O. A. (2023). Federated learning in ITS: Recent applications and open issues. IEEE Transactions on Intelligent Transportation Systems.
  6. Chong, Y. W., Yau, K. L. A., Ibrahim, N. F., & Han, J. (2024). Federated Learning for Intelligent Transport Systems: Use Cases, Open Challenges, and Opportunities. IEEE Transactions on Intelligent Transportation Systems.
  7. Wang, S., Hu, T., & Min, G. (n.d.). Federated learning in intelligent transportation systems: A survey. IEEE Transactions on ITS.
  8. Gupta, A., et al. (2020). Blockchain and AI in fare collection in public transport. Journal of Urban Mobility.
  9. Chen, Y., & Zhang, L. (2021). Federated learning approach for predictive maintenance in smart transportation systems. Journal of Intelligent Transportation Systems.
  10. Lu, Y., et al. (2021). Traffic flow prediction in smart cities based on federated learning. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS.
  11. Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020). Federated learning: Challenges, methods and future directions.
  12. Konecny, J. McMahan, H. B., Yu, F. X., Richtarik, P., Suresh, A. T., & Bacon, D. (2016). Federated learning: Towards better communication efficiency. arXiv preprint arXiv:1610.05492.
  13. Bonawitz, K., Eichner, H., Grieskamp, W., Huba, F., Ingerman, A., Jackson, N., ... & Van Der Maaten, Lecru (2019). Towards vigorous federated learning at scale: System design. In Proceedings of Machine Learning and Systems, 1, 371-382.
  14. Bagdasaryan, E., Shen, A., Sharma, A., Fong, M., Veeramachaneni, V., Sinha, N., & Shmatikov, V. (2020). How to backdoor federated learning. International Conference on Artificial Intelligence and Statistics.
  15. Fallah, A., Mokhtari, A., & Ozdaglar, A. (2020). Towards practical federated learning: fairly trading on clients' data. In Advances in Neural Information Processing Systems 33.
  16. Acar, D. A. E., Malek, A., & Kairouz, P. (2021). Federated Learning with Heterogenous Clients: A Unified View. arXiv preprint arXiv:2102.08092.
  17. Cho, Y. J., & Kim, J. (2020). Client sampling for federated learning with non-IID data in mobile edge networks. IEEE Access, 8, 195759-195769.
  18. Wang, S., Wang, T., & Liu, J. (2020). Adaptive federated learning via dot product transfer. In AAAI conference on artificial intelligence (Vol. 34, No. 04, pp. 6296-6303).
  19. Alistarh, D., Grubic, D., Li, J., Tomioka, R., & Vojnovic, M. (2017). Qsparse-local-SGD: Decentralized SGD with Quantization and Sparsification and Local Computations. In Neural Information Processing Systems, NIPS, 30.
  20. Xie, C., Koyejo, S., and Gupta, I. (2019). Asynchronous federated optimization. arXiv preprint arXiv:1903.03934.
  21. Caldas, A., Duddu, S. M. R., Li, P., Konecny, J., McMahan, H. B., & Smith, V. (2018). Extending the scale of federated learning via reducing client computational burden. arXiv preprint arXiv:1812.07279.
  22. Liu, Y., Kang, J., & Niyato, D. (2020). Vehicular networks federated transfer learning. IEEE Wireless Communications Letters, 9(10), 1730-1734.
  23. Nishio, T., & Yonetani, R. (2019). Client selection for federated learning with heterogeneous resources in mobile edge. In ICC 2019 - 2019 IEEE International Conference on Communications (ICC).
  24. Hardy, S., Henecka, M., Pfitzmann, A., & Waidner, M. (2017). Private federated learning on the plaintext. arXiv preprint arXiv:1711.08060.
  25. Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G., Davis, A., Dean, J., Devin, M., & Ghemawat, S. (2016) TensorFlow: Large-scale machine learning on heterogeneous systems. Deep learning with differential privacy. In ACM Computer Communication Security (CCS).
  26. Mo, F., Yu, Y., Li, X., & Wu, X. (2021). Private federated learning under a malicious but honest but curious model with a trust execution environment. Proceedings of the IEEE International Conference on Computer Design.
  27. Weng, J., Zhang, S., & Li, J. (2021). Towards explainable federated learning. arXiv preprint arXiv:2103.01894.
  28. Ziosi, M., et al. (2021). Bias and privacy in AI-driven public transport surveillance: A framework for responsible innovation. Ethics and Information Technology.
  29. Zhou, Z., Chen, X., Li, J. (2019). Edge intelligence: The next wave of AI. Proceedings of the IEEE, 107(8), 1738-1762.
  30. Zhang, S., Li, J., and Shi, L. Federated multi-view learning for intelligent transportation systems. IEEE Transactions on ITS.
  31. Hsieh, K., & Yang, J. (2020). The challenges of non-IID data in federated learning: A survey. arXiv preprint arXiv:2007.01439.
SciMeTech · Volume 2, Issue 1 · 2026 · ISSN 3085-5284