State of Art
In the field of machine learning operations (MLOps) and network analytics, the automation of machine learning workflows is crucial for ensuring scalability, efficiency, and reliability. In 5G networks, the Network Data Analytics Function (NWDAF) is responsible for collecting and analyzing network data to optimize performance and detect anomalies in real time.
Recent research highlights the effectiveness of machine learning techniques in identifying abnormal traffic patterns and mitigating potential Distributed Denial-of-Service (DDoS) attacks [1]. NWDAF has been demonstrated to collect real-time data from 5G networks, facilitating anomaly detection and traffic analysis [1]. Furthermore, neural networks have shown superior performance in network load prediction compared to traditional linear regression models [2], improving forecasting accuracy. However, continuous retraining of models is necessary, as they tend to become biased over time, leading to performance degradation [3].
With these advancements in mind, the main objective of this project is to implement an MLOps pipeline that, when integrated into a 5G network, serves as the foundation for NWDAF architecture. This pipeline will automate the entire lifecycle of machine learning models, from data ingestion and training to deployment and continuous monitoring.
By achieving this, the project aims to enhance the adaptability and efficiency of network analytics, ensuring real-time, data-driven decision-making in 5G environments.
References
[1]
A. Mekrache, K. Boutiba, and A. Ksentini, “Combining Network Data Analytics Function and Machine Learning for Abnormal Traffic Detection in Beyond 5G,” GLOBECOM 2023 - 2023 IEEE Global Communications Conference, Dec. 2023, https://doi.org/10.1109/globecom54140.2023.10436766.
[2]
N. Nisha, Lakshman K, and R. Kumar, “A Smart Data Analytics System Generating for 5G N/W System Via ML Based Algorithms for the Better Communications,” Apr. 2024, https://doi.org/10.1109/istems60181.2024.10560068.
[3]
Rui Cruz Ferreira et al., “Demo: Enhancing Network Performance based on 5G Network Function and Slice Load Analysis,” Jun. 2023, https://doi.org/10.1109/wowmom57956.2023.00057.