Federated DeepSeek-Assisted Traffic Prediction and Anomaly Detection in Edge-Enabled Mobile Networks
Keywords:
Federated learning, mobile edge computing, traffic prediction, anomaly detection, DeepSeek, spatial-temporal modeling, network governanceAbstract
The exponential growth of mobile data traffic and the proliferation of latency-sensitive applications demand intelligent, real-time traffic forecasting and anomaly detection mechanisms that operate close to the data source. Centralized cloud-based learning architectures incur prohibitive communication overhead, privacy vulnerabilities, and regulatory friction. This paper presents a federated learning framework that integrates DeepSeek, a large language model with strong sequence modeling and spatial-temporal reasoning capabilities, to enable collaborative traffic prediction and anomaly detection across edge nodes in mobile networks. The proposed system architecture distributes model fine-tuning across base stations and user equipment, employing parameter-efficient adaptation and communication-efficient aggregation to manage the large model footprint and heterogeneous client resources. We dissect the structural trade-offs between global prediction fidelity and communication cost, examining hierarchical aggregation, differential privacy, and resilience against adversarial perturbations. By coupling federated learning with DeepSeek, mobile operators can exploit complex long-range temporal dependencies and geographic correlations while preserving data locality, aligning with data governance frameworks, and enhancing infrastructure robustness. The analysis extends to fairness issues arising from non-identically distributed traffic patterns across diverse urban and rural regions, and discusses policy implications for cross-jurisdictional deployments. Deployment considerations address resource-constrained edge hardware, energy sustainability, and system reliability under dynamic network conditions. This study provides an interdisciplinary blueprint for a secure, scalable, and responsible socio-technical system that harnesses large-scale pre-trained models for next-generation mobile network intelligence, emphasizing the interplay between algorithmic innovation, governance, and operational sustainability.
References
1. Mao, Y., You, C., Zhang, J., Huang, K., & Letaief, K. B. (2017). A survey on mobile edge computing: The communication perspective. IEEE Communications Surveys & Tutorials, 19(4), 2322-2358.
2. Li, R., Zhao, Z., Zhou, X., Ding, G., Chen, Y., Wang, Z., & Zhang, H. (2017). The prediction analysis of cellular traffic based on big data: A deep learning approach. IEEE Access, 5, 17447-17457.
3. McMahan, B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. y. (2017). Communication-efficient learning of deep networks from decentralized data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), 1273-1282.
4. DeepSeek-AI. (2024). DeepSeek-V2: A strong, economical, and efficient mixture-of-experts language model. arXiv preprint arXiv:2405.04434.
5. Zhou, T., Niu, P., Wang, X., Sun, L., & Jin, R. (2023). One fits all: Power general time series analysis by pretrained lm. arXiv preprint arXiv:2302.11939.
6. Xu, M., Cai, S., Li, D., & Song, L. (2024). Federated large language model: A comprehensive survey. arXiv preprint arXiv:2407.10612.
7. Liu, L., Zhang, J., Song, S. H., & Letaief, K. B. (2020). Client-edge-cloud hierarchical federated learning. IEEE International Conference on Communications (ICC), 1-6.
8. Zhang, C., Zhang, H., Qiao, J., Li, Z., & Alouini, M. S. (2025). TIDES: Traffic Intelligence with DeepSeek Enhanced Spatial Temporal Prediction. IEEE Journal on Selected Areas in Communications.
9. Preuveneers, D., Rimmer, V., Tsingenopoulos, I., Spooren, J., Joosen, W., & Ilie-Zudor, E. (2018). Chained anomaly detection models for federated learning: An intrusion detection case study. Applied Sciences, 8(12), 2663.
10. Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2020). How to backdoor federated learning. International Conference on Artificial Intelligence and Statistics (AISTATS), 2938-2948.
11. Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016). Deep learning with differential privacy. ACM SIGSAC Conference on Computer and Communications Security (CCS), 308-318.
12. Li, T., Sanjabi, M., Beirami, A., & Smith, V. (2020). Fair resource allocation in federated learning. International Conference on Learning Representations (ICLR).
13. Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., ... & Zhao, S. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning, 14(1–2), 1-210.
14. Wachter, S., Mittelstadt, B., & Floridi, L. (2017). Why a right to explanation of automated decision-making does not exist in the General Data Protection Regulation. International Data Privacy Law, 7(2), 76-99.
15. Han, S., Mao, H., & Dally, W. J. (2016). Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. International Conference on Learning Representations (ICLR).
16. Schwartz, R., Dodge, J., Smith, N. A., & Etzioni, O. (2020). Green AI. Communications of the ACM, 63(12), 54-63.
17. Bonawitz, K., Eichner, H., Grieskamp, W., Huba, D., Ingerman, A., Ivanov, V., ... & Roselander, J. (2019). Towards federated learning at scale: System design. Proceedings of Machine Learning and Systems (MLSys), 1, 374-388.
18. Shen, S., Tople, S., Saxena, P., & van der Meyden, R. (2020). Auror: Defending against poisoning attacks in collaborative deep learning systems. Proceedings of the 32nd Annual Computer Security Applications Conference (ACSAC), 508-519.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Advanced Artificial Intelligence Research
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
