Blockchain-enabled collaborative anomaly detection for IoT security

Abstract

Protection of the Internet of Things (IoT) has become a significant concern due to the widespread use of IoT technologies. Conventional Intrusion Detection Systems (IDS) have challenges when used in IoT networks because of resource restrictions and complexities. Blockchain Technology (BCT) has significantly altered organizations' financial behavior and effectiveness in recent years. Data security and system stability are crucial concerns that must be tackled in blockchain systems. The study suggests a mechanism called Deep Blockchain-Enabled Collaborative Anomaly Detection (DBC-CAD) for security-focused distributed Anomaly Detection (AD) and privacy-focused BC with smart contracts in IoT networks. A Modified - Long Short-Term Memory (M-LSTM) based Deep Learning (DL) algorithm with a multi-variable optimization approach has been used for the AD approach. The multi-variable optimization technique has been used to set the hyperparameters. The Ethereum framework creates privacy-focused BC and smart contract techniques that safeguard decentralized AD engines. The proposed M-LSTM model has the highest detection rate of 99.1%. The findings show the effectiveness of the proposed systems in identifying assaults on IoT networks.