In this letter, we propose a multimodal method for improving radio frequency (RF) fingerprinting performance that uses multiple features cultivated from RF signals. Combining multiple features, including a falling transient feature that has not previously been used in RF fingerprinting studies, we aim to demonstrate that the proposed method results in improved accuracy. We show that a sparse representation-based classification (SRC) scheme can be a good platform for combining multiple features. The experimental results on RF signals acquired from eight walkie-talkies show that the RF fingerprinting accuracy of the proposed method improves significantly as the number of features increases.
In a frequency hopping spread spectrum (FHSS) network, the hopping pattern plays an important role in user authentication at the physical layer. However, recently, it has been possible to trace the hopping pattern through a blind estimation method for frequency hopping (FH) signals. If the hopping pattern can be reproduced, the attacker can imitate the FH signal and send the fake data to the FHSS system. To prevent this situation, a non-replicable authentication system that targets the physical layer of an FHSS network is required. In this study, a radio frequency fingerprinting-based emitter identification method targeting FH signals was proposed. A signal fingerprint (SF) was extracted and transformed into a spectrogram representing the time–frequency behavior of the SF. This spectrogram was trained on a deep inception network-based classifier, and an ensemble approach utilizing the multimodality of the SFs was applied. A detection algorithm was applied to the output vectors of the ensemble classifier for attacker detection. The results showed that the SF spectrogram can be effectively utilized to identify the emitter with 97% accuracy, and the output vectors of the classifier can be effectively utilized to detect the attacker with an area under the receiver operating characteristic curve of 0.99.
In Internet of Things (IoT) environments, effective public key management is crucial for managing numerous devices. RF features, primarily considered analog features within physical layer authentication by RF Fingerprinting (RFF) processes, present a novel approach to key management. In this research, we introduce a novel RF-based Public Key Generator (RF-PubKG) model that maps RF features into cryptographic sequences by incorporating a Key Generation (KeyGen) layer into the RFF model. The RF-PubKG demonstrates superior performance, achieving 97.2% accuracy at a 20dB SNR and further improving to 99.6% in noise-free conditions with a Frame Error Rate (FER) below 1%. The generated public key sets exhibit negligible correlation, with intra-key-set correlations not exceeding 0.24 and inter-model correlations falling below 0.04, highlighting the reliability of the RF-PubKG model. The integration with the Rivest–Shamir–Adleman (RSA) algorithm provides proof-of-concept for the RF-PubKG-based digital signature scheme, effectively simplifying the Certificate Authorities (CAs) management and, consequently, reducing Public Key Infrastructure (PKI) complexity. This simplification promises effective public key management within the Public Key Cryptography (PKC), thereby enhancing the efficiency of digital signature verification processes.