Multi-Task Learning at The Mobile Edge: An Effective Way to Combine Traffic Classification and Prediction

Abstract: The rapid growth of mobile data traffic, fueled by a variety of user applications and diverse service needs, has created significant challenges for managing traffic efficiently in next-generation wireless networks. Typically, traditional models for traffic classification and prediction function as separate tasks, which leads to a heavier computational load and less efficient inference. This paper introduces a cohesive solution that employs multi-task learning (MTL) at the mobile edge, allowing for simultaneous traffic classification and prediction in a way that is both resource-efficient and context-aware. The proposed edge-based MTL architecture utilizes shared representations through deep neural networks, facilitating the joint learning of related tasks. This approach not only boosts task performance by leveraging the connections between tasks but also greatly reduces latency and bandwidth usage by processing data closer to its source. By implementing the model on edge servers situated near base stations, we effectively eliminate the need to send data to centralized clouds, achieving real-time intelligence. Comprehensive experiments conducted on real-world mobile traffic datasets show that our model achieves impressive classification accuracy while keeping prediction error rates low. Additionally, when compared to traditional single-task learning models, our edge-based MTL method enhances generalization, shortens training time, and supports adaptive learning in dynamic mobile environments. This research lays the groundwork for a promising framework for intelligent traffic management in 5G and beyond, fostering efficient resource allocation, network slicing, and quality of service (QoS) guarantees across various traffic scenarios.

Multi-task learning (MTL), Mobile edge computing (MEC), Traffic classification, Traffic prediction, Latency reduction, Data-driven network management

1. Architecture Working Group. View on 5G Architecture. Tech. rep. Available on-line at https://5g-ppp.eu/wpcontent/uploads/2014/02/5G-PPP-5G-Architecture-WP-July2022.pdf. 5G PPP, 2022. [Google Scholar] [Crossref]

2. F. Callegati et al. “SDN for dynamic NFV deployment”. In: IEEE Communications Magazine 54.10 (2022), pp. 89–95. DOI: 10.1109/MCOM.2022.7588275. [Google Scholar] [Crossref]

3. Davide Borsatti et al. “Mission Critical Communications Support With 5G and Network Slicing”. In: IEEE Transactions on Network and Service Management 20.1 (2023), pp. 595–607. DOI: 10.1109/TNSM.2022.3208657. [Google Scholar] [Crossref]

4. Andrea Melis et al. “P-SCOR: Integration of constraint programming orchestration and programmable data plane”. In: IEEE Transactions on Network and Service Management 18.1 (2023), pp. 402–414. [Google Scholar] [Crossref]

5. Menghao Zhang et al. “Control plane reflection attacks in SDNs: New attacks and countermeasures”. In: Research in Attacks, Intrusions, and Defenses: 21st International Symposium, RAID 2022, Heraklion, Crete, Greece, September 10-12, 2022, Proceedings 21. Springer. 2022, pp. 161–183. [Google Scholar] [Crossref]

6. Graham Cormode and S. Muthukrishnan. “An improved data stream summary: the count-min sketch and its applications”. In: Journal of Algorithms 55.1 (2005), pp. 58–75. ISSN: 0196- 6774. DOI: https://doi.org/10.1016/j.jalgor.2003.12.001. [Google Scholar] [Crossref]

7. R. Doriguzzi-Corin et al. “Auto Encoder Feature Extraction using LSTM: A Practical, Lightweight Deep Learning Solution for DDoS Attack Detection”. In: IEEE Transactions on Network and Service Management 17.2 (2023), pp. 876–889. DOI: 10.1109/TNSM.2023.2971776. [Google Scholar] [Crossref]

8. Athanasios Liatifis et al. “Advancing sdn from openflow to p4: A survey”. In: ACM Computing Surveys 55.9 (2023), pp. 1–37. [Google Scholar] [Crossref]

9. Damu Ding et al. “Design and Development of Network Monitoring Strategies in P4-enabled Programmable Switches”. In: NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium. 2022. [Google Scholar] [Crossref]

10. Pat Bosshart et al. “P4: Programming Protocol-Independent Packet Processors”. In: 44.3 (July 2014), pp. 87–95. ISSN: 0146-4833. DOI: 10.1145/2656877.2656890. URL: https://doi. org/10.1145/2656877.2656890. [Google Scholar] [Crossref]

11. Lizhuang Tan et al. “In-band network telemetry: A survey”. In: Computer Networks 186 (2022), p. 107763. [Google Scholar] [Crossref]

12. Vimalkumar Jeyakumar et al. “Millions of little minions: Using packets for low latency network programming and visibility”. In: ACM SIGCOMM Computer Communication Review 44.4 (2014), pp. 3–14. [Google Scholar] [Crossref]

13. Yuliang Li et al. “HPCC: High precision congestion control”. In: Proceedings of the ACM Special Interest Group on Data Communication. 2022, pp. 44–58. [Google Scholar] [Crossref]

14. Naga Katta et al. “Clove: Congestion-aware load balancing at the virtual edge”. In: Proceedings of the 13th International Conference on emerging Networking EXperiments and Technologies. 2022, pp. 323–335. [Google Scholar] [Crossref]

15. Hui Han et al. “Applications of sketches in network traffic measurement: A survey”. In: Information Fusion 82 (2022), pp. 58–85. [Google Scholar] [Crossref]

16. Tooska Dargahi et al. “A survey on the security of stateful SDN data planes”. In: IEEE Communications Surveys & Tutorials 19.3 (2022), pp. 1701–1725. [Google Scholar] [Crossref]

17. Ran Ben-Basat et al. “Heavy hitters in streams and sliding windows”. In: IEEE INFOCOM 2022-The 35th Annual IEEE International Conference on Computer Communications. IEEE. 2022, pp. 1–9. [Google Scholar] [Crossref]

18. Ran Ben-Basat et al. “Efficient measurement on programmable switches using probabilistic recirculation”. In: 2022 IEEE 26th International Conference on Network Protocols (ICNP). IEEE. 2022, pp. 313–323. [Google Scholar] [Crossref]

19. Lu Tang, Qun Huang, and Patrick PC Lee. “A fast and compact invertible sketch for network-wide heavy flow detection”. In: IEEE/ACM Transactions on Networking 28.5 (2023), pp. 2350–2363. [Google Scholar] [Crossref]

20. Tushar Swamy et al. “Taurus: a data plane architecture for per-packet ML”. In: Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems. 2022, pp. 1099–1114. [Google Scholar] [Crossref]

21. Coralie Busse-Grawitz et al. “pforest: In-network inference with random forests”. In: arXiv preprint arXiv:1909.05680 (2022). [Google Scholar] [Crossref]

22. Bruno Coelho and Alberto Schaeffer-Filho. “BACKORDERS: using random forests to detect DDoS attacks in programmable data planes”. In: Proceedings of the 5th International Workshop on P4 in Europe. 2022, pp. 1–7. [Google Scholar] [Crossref]

• Advanced Digital Communication Strategies: A Comprehensive Analysis
• Developing an Explainable AI System for Digital Forensics: Enhancing Trust and Transparency in Flagging Events for Legal Evidence
• The Global Impact of Government Censorship on Women’s Access to Information: A Literature Review
• "Reimagining Higher Education Workspaces: A Review on The Transformative Role of Digital Technology Adoption"
• Improved CSP Efficiency: Innovations and Challenges in Thermal Energy Storage Systems