Physical layer authentication of off‐body channels by probabilistic neural networks. (27th May 2019)

• Record Type: Journal Article
• Authors: Raida, Zbynek
• Other Names: Soh Ping Jack guestEditor.
• Is Part Of: International journal of numerical modelling. Volume 32:Number 6(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-05-27
• Subjects: on‐body communication -- physical‐layer encryption -- probabilistic neural network -- wireless channel fingerprint; Electric networks -- Mathematical models -- Periodicals; Electronics -- Mathematical models -- Periodicals; 621.3011
• DOI: 10.1002/jnm.2628
• Languages: English
• ISSNs: 0894-3370

Abstract

Abstract: A probabilistic neural network (PNN) for the authentication of an off‐body channel of a wireless body area network is discussed in this paper. The off‐body channel is authenticated by radio‐frequency fingerprints of on‐body channels. Thanks to a relatively high stationarity of on‐body channels, a reasonable probability of radio‐frequency fingerprinting identification is reached. Radio‐frequency fingerprints are used to train a probabilistic neural network playing the role of an authentication key. The authentication process can be efficiently controlled by setting tolerance zones of the network. Functionality of the concept was proven by computer simulations.