An AI-blockchain-assisted smart agriculture framework for enabling secure and efficient data transaction: a hybrid approach

An AI-blockchain-assisted smart agriculture framework for enabling secure and efficient data transaction: a hybrid approach

연구 분야: Cryptography

학회: Knowledge and Information Systems

The integration of blockchain with the internet of things (IoT) is increasingly being explored in smart agriculture to enhance data security, reliability, and decision-making. This research addresses the critical need for a unified system that ensures secure data storage, accurate crop yield prediction, and effective attack detection within agricultural IoT networks. While previous studies have proposed various blockchain-based frameworks or deep learning models individually, few have combined both to deliver a comprehensive, high-performance solution designed for smart agriculture. This paper aims to fill that gap by introducing a Blockchain-IoT (B-IoT) framework augmented with a novel deep learning technique and a hybrid cryptographic mechanism. The system employs the interplanetary file system (IPFS) for decentralized storage, secured using a combination of digital signature algorithm (DSA) and elliptic curve cryptography (ECC). To further enhance security and reduce key generation complexity, a new equilibrium-enriched siberian (EES) algorithm is proposed, combining Siberian tiger optimization (STO) and equilibrium optimization (EO) strategies. For intelligent analysis, a modified capsule neural network (M-CapNN) is implemented to predict crop yield and detect attacks early. The Experimental results demonstrate the superiority of the proposed framework, achieving an accuracy of 99.97%, sensitivity of 99.50%, and F1-score of 99.62%, thereby outperforming existing state-of-the-art methods. These findings suggest that the proposed B-IoT system offers a robust and efficient solution for secure and intelligent agricultural monitoring.