Evaluating Drift Detection Methods for Failure Prediction in 5G Network Resource Management
Resumo
In dynamic cloud environments, ensuring accurate and adaptive resource allocation is essential to maintain 5G network performance. This study compares concept drift detection strategies to improve fault tolerance and resource efficiency using the 5G3E dataset, which encompasses abrupt, gradual, and recurring changes in CPU and network demand. An LSTM model was deployed to predict resource usage, and three drift detection techniques—Page-Hinkley (PHT), ADWIN, and KSWIN—were applied to the prediction residuals. Experimental results show that PHT captured 50% of the CPU drifts and 66% and KSWIN, conversely, detected every drift event (100% recall) at the expense of numerous false positives. These findings highlight the trade-off between sensitivity and precision in dynamic 5G scenarios, underscoring the need to select a drift detection method based on acceptable levels of missed detections versus false alarms.
Referências
Barbero, F., Pendlebury, F., Pierazzi, F., and Cavallaro, L. (2022). Transcending transcend: Revisiting malware classification in the presence of concept drift. In 2022 IEEE Symposium on Security and Privacy (SP), pages 805–823. IEEE.
Barros, R. S. M. and Santos, S. G. T. C. (2018). A large-scale comparison of concept drift detectors. Information Sciences, 451-452:348–370.
Cabello-López, T., Cañizares-Juan, M., Carranza-García, M., Garcia-Gutiérrez, J., and Riquelme, J. C. (2022). Concept drift detection to improve time series forecasting of wind energy generation. In International Conference on Hybrid Artificial Intelligence Systems, pages 133–140. Springer.
Cerqueira, V., Gomes, H. M., Bifet, A., and Torgo, L. (2023). Studd: A student–teacher method for unsupervised concept drift detection. Machine Learning, 112(11):4351–4378.
Gama, J., Medas, P., Castillo, G., and Rodrigues, P. (2004). Learning with drift detection. In Proceedings of the Brazilian Symposium on Artificial Intelligence (SBIA), pages 286–295, São Luís, Brazil. Springer.
Gonçalves, G. E., Santos, G. L., Ferreira, L., Rocha, É. d. S., de Souza, L. M., Moreira, A. L., Kelner, J., and Sadok, D. (2020). Flying to the clouds: the evolution of the 5G radio access networks. In The Cloud-to-Thing Continuum: Opportunities and Challenges in Cloud, Fog and Edge Computing, pages 41–60.
Gudepu, V., Abbas, A., Doe, J., and Smith, J. (2024). The drift handling framework for open radio access networks: An experimental evaluation. Journal of Network and Systems Management, 32:45–63.
Hu, P. and Zhang, J. (2020). 5G-enabled fault detection and diagnostics: How do we achieve efficiency? IEEE Internet of Things Journal, 7(4):3267–3281.
Kassi, M. and Hamouda, S. (2024). Ran virtualization: How hard is it to fully achieve? IEEE Access.
Mehmood, T., Latif, S., Jamail, N. S. M., Malik, A., and Latif, R. (2023). LSTMDD: An optimized LSTM-based drift detector for concept drift in dynamic cloud computing. PeerJ Computer Science. Manuscript submitted, under review.
Phung, C.-D., Ruba, S. B., and Secci, S. (2022). An open dataset for beyond-5G data-driven network automation experiments. In 2022 1st International Conference on 6G Networking (6GNet), pages 1–4. IEEE.
Saxena, D. and Singh, A. K. (2022). OFP-TM: An online vm failure prediction and tolerance model towards high availability of cloud computing environments. The Journal of Supercomputing, 78:8003–8024.
Zhang, Z. and Zhang, H. (2024). An online transfer learning model for intrusion detection using ft-transformer and kswin-driven concept drift detection mechanism. In 2024 5th International Conference on Computer Engineering and Application (ICCEA), pages 128–131. IEEE.
