Skip to content
Conference Open access

Robust Radio Frequency Fingerprint Identification Under Phase Noise and Doppler Shifts

Sep 2026 · 2026 10th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) · pp. 528-535 · 0 citations · 16 references

TL;DR

A self-supervised contrastive RFFI framework that learns robust impairment-invariant, device-discriminative representations from unlabelled RF signals is proposed, demonstrating enhanced robustness compared to conventional CNN, ResNet and contrastive RFFI baselines.

Abstract

Radio frequency fingerprint identification (RFFI) is susceptible to physical impairments that distort transmitter-specific features, especially phase noise and Doppler shifts. This paper proposes a self-supervised contrastive RFFI framework that learns robust impairment-invariant, device-discriminative representations from unlabelled RF signals. The method creates a number of views of the same transmission under controlled phase-noise and Doppler conditions, which are considered as positives against signals from different transmitters as negatives. Common RF encoder and projection head is trained with an InfoNCE loss and fine-tune them in a supervised way for transmitter identification. The evaluation includes clean signals, individual impairment severities, joint phase-noise and Doppler conditions, and augmentation ablations. The evaluation is conducted using session-disjoint transmitter data under controlled phase-noise and Doppler conditions, including both individual and joint impairment scenarios. The proposed method achieves 96.1% accuracy in clean conditions, 91.2% accuracy in severe phase noise, 92.4% accuracy in severe Doppler, and 88.7% accuracy in severe joint impairment. These results demonstrate enhanced robustness compared to conventional CNN, ResNet and contrastive RFFI baselines, while preserving device-discriminative information under different wireless physical conditions.

Read PDF

Similar papers

2026

Index Modulation Leveraging RF Fingerprints: System Design and Prototype Validation

This letter exploits radio frequency (RF) fingerprints as a new information-bearing index domain beyond their conventional use in device identification or authentication, and proposes two index-only schemes: RF fingerprint modulation (RFFM), which conveys index bits via individual RF-chain fingerprints, and generalized...

Wei-Dong Liu, Nan Zhao, Jing Zhu et al. · 0 citations
Open access 2026

Lightweight Memory Network for Drone Type RF Fingerprint Identification in Dynamic Environments

Radio frequency fingerprint identification (RFFI) is a crucial physical layer security technique for unauthorized drone identification. However, current drone RFFI methods are primarily designed for line-of-sight (LOS) video transmission or remote control signals while suffering significant performance degradation unde...

Fei-Fei Zhu, Fu-Hui Zhou, Qi-Hui Wu · 0 citations
Open access 2026

Semi-Supervised RF Fingerprinting: Contrastive Representation Alignment and Unknown-Class Discovery

The proposed Semi-Supervised RF Fingerprinting: Contrastive Representation Alignment and Unknown-Class Discovery (SRCU) framework enables the joint identification of known emitters and discovery of multiple unknown classes within a single training process.

Ke Wang, Yi-Ting Gao, Zhi-Ying Wang et al. · 0 citations
Sep 2026

SFPF: Spatio-Frequency Polarization Fingerprint for Anomalous Wireless Device Detection

Periodic inspection of deployed wireless devices is necessary because unauthorized hardware replacement may preserve communication functions, credentials, and logical identity, making anomalous devices difficult to detect. Such inspections are conducted under controlled measurement conditions to verify that each device...

Xiao-Xuan Huang, Jin-Long Xu, Da-Yu Shen et al. · 0 citations
Open access 2026

EJ-ANet: A Lightweight Envelope-Jitter Asymmetric Network for Drone Radio Frequency Fingerprint Identification

Drone radio frequency fingerprint identification (RFFI) provides a passive solution for fine-grained drone recognition. However, time-frequency image-based methods introduce additional preprocessing costs, while raw in-phase/quadrature (IQ) networks remain limited in modeling complementary temporal variations. This let...

Shen-Feng Ye, Dan-Yang Jiang, Zi-Hang Qi · 0 citations
Preprint Oct 2026

From Identification to Discovery: Holistic RF Fingerprinting via Learning-Aided Quantization

Radio frequency fingerprint identification (RFFI) enables device identification and authentication by exploiting transmitter-specific hardware impairments embedded in received RF signals. Existing RFFI methods are typically designed for a specific operating regime, such as closed-set identification, open-set recognitio...

Omer Hazan, Gil Paryanti, Nir Shlezinger · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.