Blockchain-Enabled Cross-Layer Radio Frequency Fingerprinting Identification with Machine Learning for IIoT

Zhiwei Chen,Hong Wen,Wenjing Hou,Yixin Jiang,Liang Chen,Runhui Zhao,Xinyu Hou
DOI: https://doi.org/10.1109/mnet.002.2200635
IF: 10.294
2023-01-01
IEEE Network
Abstract:Motivated by rising voices of light weight high security for the Industrial Internet of Things (IIoT), in this article, we propose a cross-layer authentication scheme by taking advantage of both the radio frequency fingerprinting (RFF) and the blockchain trust. Due to the mobility of terminal devices in the IIoT network, the wireless terminal devices often move from one region to another region, which causes traditional RFF identification scheme to not be able to be adopted. In this article, the RFFs of the wireless terminal devices that are easy to invade are extracted and learned by the first access edge device. Then the learned model is shared with other edge devices by aiding with blockchain. This way, IIoT network identifies wireless terminal nodes with high security by using the unique structure of the transmitter waveform without adding any burden while repetitive training is avoided. Besides, an evaluation metric has been proposed to evaluate the performance of robust RFF identification in practical mobile scenarios. Lastly, this article further presents several research directions for RFF identification in practical mobile scenarios.
What problem does this paper attempt to address?