Authenticated Key Agreement Scheme for Securing IoMT-Based Consumer Smart Healthcare Using Reinforcement Learning-as-a-Service

Smart healthcare security is a critical research area today. Authentication is a fundamental security feature, especially in systems dealing with sensitive healthcare data. The complexity of current consumer-centric smart healthcare systems and the increasing number of security breaches highlight the need for robust authentication methods. This study proposes reinforcement learning as a service-based authentication scheme that integrates risk assessment and authentication scheme selection. The proposed scheme aims to provide adaptive authentication, capable of addressing adaptive authentication features and offering design-level security. The proposed authentication scheme utilizes reinforcement learning (RL) as a service to enhance the security and functionality of traditional authentication methods from a consumer-centric smart healthcare perspective. This scheme represents a unique approach in the field of authentication mechanisms. Furthermore, we introduce a novel methodology that uniquely helps RL to be used as a service for a cryptographic authentication scheme, while also leveraging trust establishment to enhance authentication efficiency. The scheme’s performance is evaluated based on communication, computation, and security parameters. Additionally, an experiment is employed to assess its practical performance.