IoT Sensor Network and Ethereum-based Blockchain for Enhanced Pest Identification and Water Management in Vineyards

Abstract

The emergent high-speed growth of precision agriculture requires solid frameworks that will improve sustainability, efficiency, and transparency in managing vineyards. The proposed research suggests a system that uses IoT sensor networks, machine learning models, and Ethereum-based blockchain to solve two important problems: pest identification and the optimization of water resources. The IoT layer was used to implement soil moisture, climate, and imaging sensors to gather real-time data about the vineyard. At the edge, preprocessing and deep learning algorithms were used with a blockchain-enabled Convolutional Neural Network (CNN) to provide correct pest detection. At the same time, the timing of irrigation was automated by IoT-enabled soil moisture monitoring, which greatly decreased the amount of wasted water. This ensured integrity of data and trust in the farmers as the Ethereum blockchain layer offered immutable storage, smart contracts to make decisions, and secure events logging. The results of the experiments revealed that the proposed system had a 95.8% pest detection accuracy, which was higher than the traditional and baseline machine learning methods. The efficiency of water management increased too by reducing water consumption by 55.2 % and doubling the crop productivity by 26.8 %t. Moreover, the blockchain application has scored high security of 0.93, confirming that it is a reliable solution even with moderate latency overhead. In this study, the authors emphasize the opportunities of converging the IoT with blockchain and transforming viticulture through sustainable practices, data safety, and efficiency.