ML-Enabled Eavesdropper Detection in Beyond 5G IIoT Networks

2025-05-05

AI Summary

To address the scalability and deployment complexity limitations of conventional cryptography in B5G industrial IoT (IIoT), this work proposes a decentralized, real-time eavesdropper detection method grounded in physical-layer security (PLS). Leveraging lightweight physical features—including channel state information (CSI), node location, and transmit power—we design a hybrid AI model integrating a deep convolutional neural network (DCNN) with a random forest classifier. This is the first systematic validation of PLS-based eavesdropping detection in heterogeneous B5G IIoT environments. Our approach achieves near-perfect detection accuracy (~100%) and zero false positives—substantially outperforming cryptographic alternatives. By bridging artificial intelligence and physical-layer security, this work establishes a low-overhead, highly robust security enhancement paradigm for large-scale industrial wireless networks.

Technology Category

• Machine Learning: Hardware-aware ML
• Philosophy and Ethics of AI: Privacy & Security
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Advanced fifth generation (5G) and beyond (B5G) communication networks have revolutionized wireless technologies, supporting ultra-high data rates, low latency, and massive connectivity. However, they also introduce vulnerabilities, particularly in decentralized Industrial Internet of Things (IIoT) environments. Traditional cryptographic methods struggle with scalability and complexity, leading researchers to explore Artificial Intelligence (AI)-driven physical layer techniques for secure communications. In this context, this paper focuses on the utilization of Machine and Deep Learning (ML/DL) techniques to tackle with the common problem of eavesdropping detection. To this end, a simulated industrial B5G heterogeneous wireless network is used to evaluate the performance of various ML/DL models, including Random Forests (RF), Deep Convolutional Neural Networks (DCNN), and Long Short-Term Memory (LSTM) networks. These models classify users as either legitimate or malicious ones based on channel state information (CSI), position data, and transmission power. According to the presented numerical results, DCNN and RF models achieve a detection accuracy approaching 100% in identifying eavesdroppers with zero false alarms. In general, this work underlines the great potential of combining AI and Physical Layer Security (PLS) for next-generation wireless networks in order to address evolving security threats.

Problem

• Detect eavesdroppers in B5G IIoT networks using ML/DL
• Evaluate ML models for classifying legitimate vs malicious users
• Achieve high accuracy in eavesdropper detection with zero false alarms

Innovation

• ML/DL models detect eavesdroppers using CSI data
• DCNN and RF achieve near 100% detection accuracy
• AI enhances Physical Layer Security in B5G networks