Abstract

Ubiquitous sustainable computing systems are deployed for providing complex application services and processing for real-time applications. Data exchange and sharing are confidential for administering user security and privacy. The need for confidentiality is one of the significant obstacles to ubiquitous computing areas like human–machine interfaces and data security. It is crucial in ubiquitous computing to protect user and system security, privacy, and safety. Hence, this article introduces a Linearized Mutual Security Scheme (LMSS) to prevent data leakage and security breaches. The scheme provides mutual authentication based on security requirements depending upon the ubiquitous application. The user credentials are pre-shared with the service providers for determining the security level and certificate sharing. This sharing is updated at fixed intervals for deciding on authentication revocation and strength updates. In this scheme, federated learning is exploited for identifying user security demands and levels administered by different users. Based on the learning recommendations, the key authentication factors are updated in the certificate, preventing anonymous and tracking sessions in the sharing interval. Therefore, the proposed scheme achieves fair anomaly detection of 14.47%, reduces data leakage rate by 13.68%, and halts sessions by 10.26%. Besides, it improves the sharing ratio of 6.92% and application responses.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call