Work

Improving intrusion detection in O-RAN with synthetic data generation: A GAN and SMOTE approach

Year: 2025

Type: article

Abstract: In network security, the issue of imbalanced data, where certain classes are underrepresented, poses a significant challenge for intrusion detection systems (IDS), as machine learning algorithms tend to prioritize the majority class, resulting in poor detection of minority classes and reduced overall model accuracy. Accurately identifying and utilizing these rare intrusions (minority classes), despite their lower prevalence in network traffic, is essential for predicting and mitigating emerging threats. This paper presents an advanced IDS designed for Open Radio Access Networks (O-RAN) to address class imbalance by employing a combination of sampling techniques and Generative AI (GenAI), while comparing the effectiveness of single learners and ensemble classifiers. The system incorporates synthetic data generated through the Synthetic Minority Oversampling Technique (SMOTE) and Generative Adversarial Network (GAN) variants, and its performance was evaluated on two datasets: CICEVSE2024 and real-world Open RAN datasets. While SMOTE improved the accuracy of weaker classifiers, such as Naive Bayes (NB), by 10%, its impact on other classifiers was less pronounced. Multiple key performance indicators (KPIs), including accuracy and F1-score, were used to evaluate the classifiers, providing insights for optimizing IDS performance to secure Open RAN networks.

Source: Telematics and Informatics Reports

DOI: 10.1016/j.teler.2025.100269

Authors: Emmanuel N. Amachaghi, Sulyman Age Abdulkareem, Chuan Heng Foh, De Mi, Mohammad Shojafar

Institutions: University of Surrey, Birmingham City University

Language: English

Topic: Network Security and Intrusion Detection

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $1,790

Funders: Engineering and Physical Sciences Research Council, Innovate UK, Shenzhen Science and Technology Innovation Program