Towards Scalable and Channel-Robust Radio Frequency Fingerprint Identification for LoRa

Abstract

Radio frequency fingerprint identification (RFFI) is a promising device authentication technique based on the transmitter hardware impairments. In this paper, we propose a scalable and robust RFFI framework achieved by deep learning powered radio frequency fingerprint (RFF) extractor. Specifically, we leverage the deep metric learning to train an RFF extractor, which has excellent generalization ability and can extract RFFs from previously unseen devices. Any devices can be enrolled via the pre-trained RFF extractor and the RFF database can be maintained efficiently for allowing devices to join and leave. Wireless channel impacts the RFF extraction and is tackled by exploiting channel independent feature and data augmentation. We carried out extensive experimental evaluation involving 60 commercial off-the-shelf LoRa devices and a USRP N210 software defined radio platform. The results have successfully demonstrated that our framework can achieve excellent generalization abilities for device classification and rogue device detection as well as effective channel mitigation.

Disentangled Representation Learning for RF Fingerprint Extraction under Unknown Channel Statistics

“…where I(x; x) and I(y; x) respectively quantify the amount of information that the background signal x contains about that the original signal x and the identity y, and ≥ 0 is a hyperparameter that controls the amount of device-relevant information that remains in x. To facilitate the subsequent development, we further relax the problem in (14) and convert it into an unconstrained problem by using a quadratic penalty [42] as follows:…”

“…In the results depicted in the next section, we set the hyper-parameters as λ = 0.5, α = 10, and β = 10. We also set the hyper-parameter in the information constraint in (14) to be 0, i.e., = 0. As in the previous work [25], we set the radius of the HP to be δ = 10.…”

“…In our previous work [13], a softmax-based metric learning approach was adopted for enabling device authentication by strengthening the generalizability of the DL-based RFFs, such that unknown devices can be recognized even with the existence of unknown device aging. Simliarly, in [14], a metric learning approach based on triplet loss was proposed to enable unknown device authentication. Thanks to the ability of DNNs to learn the RFF features themselves, the classification performance of these DL-based RFFs is noticeably better than that obtained with handcrafted RFFs.…”

“…In real-world IoT networks, collecting data representative of all possible channel conditions is prohibitively expensive if not impossible. One possible approach for avoiding this overfitting in the DL-based RFFs is to apply data augmentation (DA) techniques for training the RFF extractor [14]- [16]. Despite certain improvements, DA-based methods still heavily rely on the specific channel models as prior knowledge.…”

Conventional Neural Network-Based Radio Frequency Fingerprint Identification Using Raw I/Q Data

“…By analysing the subtle differences of transmitters' RF fingerprints, RF fingerprint-based learning models could distinguish varieties of devices, thereby being difficult to clone and fake [1][2][3][4]. These subtle differences of transmitters are normally hard to be identified, and artificial intelligence (AI) methods could be appropriate to mitigate this problem [5][6][7].…”