A Demonstration of Self-Adaptive Jamming Attack Detection in AI/ML Integrated O-RAN

TL;DR

The paper tackles jammer and interference threats in 5G O-RAN by designing SAJD, a self-adaptive jammer-detection framework that operates in a closed-loop AI/ML pipeline. It integrates Labeler rApp, Training Manager rApp, and an ML-based interference-detection xApp under ClearML to autonomously label data, retrain models, and deploy updated detectors without service disruption. The approach leverages live KPI streams from non-RT and near-RT RICs and uses a three-layer dense neural network for real-time binary decisions on interference. In O-RAN testbed experiments, SAJD outperforms offline, manually labeled state-of-the-art detectors under unseen interference, demonstrating robust adaptability and scalability for future RAN interference management, including potential 6G readiness.

Abstract

The open radio access network (O-RAN) enables modular, intelligent, and programmable 5G network architectures through the adoption of software-defined networking, network function virtualization, and implementation of standardized open interfaces. However, one of the security concerns for O-RAN, which can severely undermine network performance, is jamming attacks. This paper presents SAJD- a self-adaptive jammer detection framework that autonomously detects jamming attacks in AI/ML framework-integrated ORAN environments without human intervention. The SAJD framework forms a closed-loop system that includes near-realtime inference of radio signal jamming via our developed ML-based xApp, as well as continuous monitoring and retraining pipelines through rApps. In this demonstration, we will show how SAJD outperforms state-of-the-art jamming detection xApp (offline trained with manual labels) in terms of accuracy and adaptability under various dynamic and previously unseen interference scenarios in the O-RAN-compliant testbed.

Figures (3)

• Figure 1: SAJD Closed-Loop Framework.
• Figure 2: Labeler rApp prediction performance for different scenarios.
• Figure 3: Comparison of interference detection accuracy across windows of sequential scenes (1a–2f).