Radio frequency (RF) fingerprinting extracts fingerprint features from RF signals to protect against masquerade attacks by\nenabling reliable authentication of communication devices at the â??serial numberâ? level. Facilitating the reliable authentication of\ncommunication devices are machine learning (ML) algorithms which find meaningful statistical differences between measured\ndata. The Generalized Relevance Learning Vector Quantization-Improved (GRLVQI) classifier is one ML algorithm which has\nshown efficacy for RF fingerprinting device discrimination. GRLVQI extends the Learning Vector Quantization (LVQ) family of\nâ??winner take allâ? classifiers that develop prototype vectors (PVs) which represent data. In LVQ algorithms, distances are\ncomputed between exemplars and PVs, and PVs are iteratively moved to accurately represent the data. GRLVQI extends LVQ\nwith a sigmoidal cost function, relevance learning, and PV update logic improvements. However, both LVQ and GRLVQI are\nlimited due to a reliance on squared Euclidean distance measures and a seemingly complex algorithm structure if changes are\nmade to the underlying distance measure. Herein, the authors (1) develop GRLVQI-D (distance), an extension of GRLVQI to\nconsider alternative distance measures and (2) present the Cosine GRLVQI classifier using this framework. To evaluate this\nframework, the authors consider experimentally collected Z-wave RF signals and develop RF fingerprints to identify devices.\nZ-wave devices are low-cost, low-power communication technologies seen increasingly in critical infrastructure. Both classification\nand verification, claimed identity, and performance comparisons are made with the new Cosine GRLVQI algorithm. The\nresults show more robust performance when using the Cosine GRLVQI algorithm when compared with four algorithms in the\nliterature. Additionally, the methodology used to create Cosine GRLVQI is generalizable to alternative measures.
Loading....