Abstract

The widespread deployment of Internet of Things (IoT) devices in smart healthcare, industrial automation, and intelligent transportation systems has widened the attack surface of distributed systems. Resource-constrained IoT nodes remain vulnerable to identity spoofing, data tampering, replay attacks, and denial-of-service intrusions due to limited computational and storage capacity. This paper proposes an Adaptive IoT Security Algorithm (AISA) that combines selected lightweight cryptography with a blockchain-based trust management framework to achieve scalable, privacy-preserving architectures. The framework dynamically selects cryptographic configurations based on device capability scores, whereas a permissioned blockchain provides decentralized authentication, an unchangeable audit log, and secure transaction validation. The optimized lightweight cryptography mechanism, combined with selective consensus based on PBFT, reduces detection latency to 36.4 milliseconds while maintaining a throughput of 248 transactions per second in large-node IoT environments. For operational cycles, energy consumption is only 3.2 J, and the blockchain overhead is only 9.7%, making it an efficient way to handle consensus. The proposed framework has a Scalability Index of 0.81, indicating excellent performance under increasing network size and load. The adaptive model is more efficient in terms of computation and resource use than the static, standalone security strategies and offers strong resistance to replay and impersonation attacks. The findings establish that the resource-aware lightweight cryptography and blockchain-based distributed trust provide a trade-off between the security strength, scalability, and computational sustainability that is balanced, and therefore, the architecture can be applied in large-scale IoT deployments.