Blockchain Based Secure Data Sharing in Precision Agriculture: a Comprehensive Methodology Incorporating Deep learning and Hybrid Encryption Model

Abstract

The precision agriculture discipline swiftly adopted blockchain as a key technology in numerous applications. From just smart farms to an internet of smart farms in precision farming, the Internet of Things (IoT) and blockchain is going to boost crop yield in precision agriculture. This proposed methodology presents a comprehensive approach for secure data sharing in precision agriculture. It integrates advanced techniques through multiple layers, including Data Collection, Data Preprocessing, Intelligent Analysis, Security, and Blockchain. IoT sensors collect data on soil moisture, temperature, humidity, crop health, and weather conditions. Preprocessing involves removing outliers, normalizing values, and extracting relevant features, then the required features are selected using the hybrid Sand Cat with Fire Hawk Algorithm (HSCFHA) which is the combination of standard Fire Hawk Optimization (FHO) and Sand Cat Swarm Optimization (SCSO). Deep learning models like Capsule Neural Network (CapsNet), Recurrent Neural Network (RNN), and Bi-directional Long Short-Term Memory (Bi-LSTM) networks provide valuable insights for prediction, classification, and anomaly detection. The blockchain layer establishes a decentralized and tamper-resistant ledger for transparent and immutable data transactions. Smart contracts automate and enforce data sharing rules. By incorporating optimized clustering, deep learning models, hybrid encryption, and blockchain technology, this framework empowers precision agriculture while ensuring secure and efficient data sharing with 98.31% of accuracy.

Keywords:
Precision agriculture; blockchain; Internet of Things; Security; Capsule Neural Network; Bi-directional Long Short-Term Memory

HIGHLIGHTS

This proposed methodology presents a comprehensive approach for secure data sharing in precision agriculture.

Deep learning models like Capsule Neural Network (CapsNet), Recurrent Neural Network (RNN), and Bi-directional Long Short-Term Memory (Bi-LSTM) networks provide valuable insights for prediction, classification, and anomaly detection.

This framework empowers precision agriculture while ensuring secure and efficient data sharing with 98.31% of accuracy.