Zero-trust blockchain-enabled framework for scalable and secure IoT networks

Abstract

The rapid expansion of IoT device implementation in environments such as smart cities, factories, and marine industries has introduced significant challenges in security and scalability. Existing blockchain-based IoT security solutions primarily rely on implicit trust to authenticate devices, leaving networks vulnerable to insider attacks, where pre-authenticated devices are compromised to gain unauthorized access. Moreover, the exponential growth in IoT devices exacerbates scalability bottlenecks, overwhelming traditional solutions. In this paper, we propose the Zero-Trust Blockchain-Enabled Framework for Scalable and Secure IoT Networks (ZT-BlocIoT), a comprehensive solution grounded in zero-trust principles, blockchain technology, and AI-based optimization. The framework employs dynamic trusted gateways for continuous device authentication and integrates a Deep Reinforcement Learning (DRL)-driven sharding mechanism to enhance blockchain scalability. The DRL approach dynamically balances shard workloads, minimizes cross-shard transactions, and mitigates adaptive collusion attacks. Evaluation results confirm ZT-BlocIoT’s superior performance, achieving 2500 TPS at 700 nodes, a 14 % increase over competing frameworks. Latency remains at 42 ms, significantly lower than alternatives that reach up to 65 ms. The framework reduces cross-shard transactions, improves node trust evaluation by 41 %, and maintains 85 % throughput even with 20 % malicious nodes, demonstrating strong resilience and security efficiency. These findings validate its AI-driven optimization, enhancing blockchain scalability, security, and real-time adaptability, making it a highly effective solution for secure IoT ecosystems.