Scalable IoT Security Framework Leveraging Ensemble Learning for Attestation

Attestation involves providing verifiable evidence to support claims about the characteristics of a target device. Devices that manage sensitive data-such as wearables, Industrial IoT devices, and servers-are vulnerable to various threats, including edge-based attacks, denial-of-service (DoS) attacks, counterfeiting, and firmware tampering. Attestation plays a crucial role in ensuring the integrity of firmware and configurations, authenticating genuine hardware, and verifying that composite device assemblies remain unaltered. However, a review of existing solutions highlights significant deficiencies, particularly in comprehensive lifecycle management and support for large-scale networks. This paper introduces a novel attestation method based on an optimized ensemble learning, which achieves a predictive accuracy of 93%, addressing these gaps and enhancing security across the entire device lifecycle.