Adversarial-Resilient RF Fingerprinting: A CNN-GAN Framework for Rogue Transmitter Detection

2025-10-07

AI Summary

To address the challenge of rogue transmitters maliciously imitating legitimate devices’ radio-frequency fingerprints (RFFs) to evade detection, this paper proposes a CNN-GAN adversarial framework. A convolutional neural network (CNN) extracts hardware-specific RFF features from in-phase/quadrature (IQ) signals, while a generative adversarial network (GAN) synthesizes realistic counterfeit IQ signals to simulate adversarial attacks—a novel application in RFF-based authentication. A softmax-based adaptive thresholding mechanism is further introduced for binary authenticity classification. The method significantly enhances the robustness and adversarial defense capability of RFF systems. Experiments conducted on ten ADALM-PLUTO software-defined radio (SDR) devices demonstrate successful discrimination among seven legitimate transmitters and two rogue emitters; the tenth device is used for optimal threshold calibration, achieving both false positive and false negative rates below 5%.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Intelligent Robots: Cognitive Robotics

Application Category

• Security and Privacy: Tracking, profiling, and countermeasures against them
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Radio Frequency Fingerprinting (RFF) has evolved as an effective solution for authenticating devices by leveraging the unique imperfections in hardware components involved in the signal generation process. In this work, we propose a Convolutional Neural Network (CNN) based framework for detecting rogue devices and identifying genuine ones using softmax probability thresholding. We emulate an attack scenario in which adversaries attempt to mimic the RF characteristics of genuine devices by training a Generative Adversarial Network (GAN) using In-phase and Quadrature (IQ) samples from genuine devices. The proposed approach is verified using IQ samples collected from ten different ADALM-PLUTO Software Defined Radios (SDRs), with seven devices considered genuine, two as rogue, and one used for validation to determine the threshold.

Problem

• Detecting rogue wireless transmitters using RF fingerprinting
• Defending against adversarial imitation of device RF characteristics
• Authenticating devices through hardware imperfection analysis

Innovation

• CNN framework detects rogue transmitters
• GAN emulates attacks using genuine IQ samples
• Softmax thresholding validates device authenticity