An AI-Enabled Hybrid Lightweight Authentication Scheme for Intelligent IoMT Based Cyber-Physical Systems

Abstract

In the era of smart healthcare, Internet of Medical Things (IoMT) based Cyber-Physical Systems (CPS) plays an important role, while accessing, monitoring, assessing, and prescribing patients ubiquitously. Efficient authentication and secure data transmission are the influential impediments of these networks to be addressed to maintain credence amidst clients, healthcare specialists, pharmacologists, and other associated entities. To address the authentication and data privacy issues in smart healthcare, in this paper, we proposed a hybrid lightweight authentication scheme leveraging supervised machine learning (SML) technique followed by Cryptographic Parameter Based Encryption and Decryption (CPBE&D) scheme to ensure the validation of legal patient wearable devices with secure transmission over the wireless communication channel. To achieve better results, we enabled the decentralized authentication of legitimate patient wearable devices to minimize computation cost, authentication time, and communication overhead with the help of SML technique to predicate and forward authentication attributes of patient wearable devices to the next concerned trusted authority, when it is shifted from region to another region. Simulation upshots of the SML and CPBE&D scheme exhibit extraordinary security features with cost-effective validation of legal patient wearable devices accompanied by worthwhile communication functionalities compared with the predecessor work.