Improving IoT Device Authentication with Lightweight AI Algorithms for Restricted Contextual Environment

Abstract

Like CPU capability, memory, and battery life, the ecosystem of Abstract The fast development of the Internet of Things (IoT) connects a wide range of items often with little processing capacity. Conventional authentication techniques provide tremendous challenges as they may need complex computations beyond of these devices' reach. This paper investigates under constrained settings the use of lightweight artificial intelligence (AI) systems to enhance IoT device authentication. Among light-weight artificial intelligence solutions that balance security and efficiency are decision trees, support vector machines (SVM), and k-nearest neighbors (k-NN), which let IoT devices safely check one another without taxing their limited resources. By including these technologies into authentication procedures IoT systems may achieve high security standards without sacrificing speed. Reviewing many implementations, the paper looks at the primary benefits of lightweight artificial intelligence in enhancing authentication processes and explores important challenges and future directions for using AI-driven solutions in IoT security.