Secure Communication Framework for IoT Devices Using Blockchain-Integrated Lightweight Encryption

Abstract

The ubiquitous deployment of the Internet of Things (IoT) systems in the areas of essential industries, such as industrial automation, healthcare, and intelligent urban infrastructure, have essentially changed the contemporary data exchange. Nevertheless, heterogeneous and decentralized IoT ecosystems pose extreme security risks especially in data confidentiality, authentication of devices, and data provenance. The existing cryptographic primitives (like the Advanced Encryption Standard (AES) and the Rivest Shamir Adleman (RSA)) have computational and energy consumption that is beyond the operational limitations of resource-constrained edge devices. To mitigate these systemic shortcomings, this study proposes a new highly scalable secure communication system, which is symbiotically coupled with lightweight cryptographic algorithms, and permissioned edge-blockchain system. The framework suggested uses the ASCON and SPECK lightweight ciphers to securely encrypt a payload in a highly efficient manner, hybridized with Elliptic Curve Cryptography (ECC) to securely exchange keys with minimal overhead. This cryptographic layer is mathematically anchored to a distributed ledger based on the Hyperledger Fabric and placed at the network edge. The architecture, based on an optimized Practical Byzantine Fault Tolerance (PBFT) consensus mechanism, using queuing theory modeling, ensures non-mutability of transactions and autonomous access control through smart contracts without overloading the sensory nodes. Large-scale empirical tests based on NS-3 to model networks, Contiki Cooja to profile constrained devices, and the CICIoT2023 dataset to test network resilience to intrusions show significant performance benefits. The architecture consumes less cryptography power of up to 39.2 mW when using ASCON-128a and the end-to-end transaction latency of up to 87 ms on average, and a throughput of 550 transactions per second. Anomaly detection models built on machine learning and implemented at the edge had an accuracy rate of 99.89 percent in neutralizing advanced vectors prior to ledger committal. The paper adds to the verifiable, low-overhead architecture blueprint to ensure the security of next-generation IoT deployments against changing cyber-physical threats.