Ethical Design of Edge-based Smart Irrigation System by use Case Analysis

Smart irrigation systems are based on Internet of Things (IoT), Artificial Intelligence (AI), Machine Learning (ML) and big data technologies, and they have the advantages of optimizing the water consumption, while ensuring the maximum crop yield. It is essential to ensure that the deployed smart irrigation systems are ethically designed, to build the trust of the users. Trust is established by compliance to agreed-upon ethical rules, regulations, and policies, along with technological aspects in the design of the system. Privacy and security, form the technological basis for ethical IoT design. The goal of this paper is to present an approach for designing a smart irrigation system that is ethical. The steps involved in the proposed approach are, definition of system architecture and identifying its elements, analyzing the security and privacy vulnerabilities by use case analysis, application of the existing solutions, and finally, the evaluation on how ethical is the system design. Although there is literature available on ethical design of IoT in general, the lack of use cases for smart irrigation vertical, motivates this paper.

I. Introduction

Usage of Internet of Things (IoT) technology, involving sensors for obtaining the information of the soil, weather, and crop, etc., for automating the watering, is called smart irrigation. The objective is to increase crop yield, while reducing the consumption of power and water [1]. It also helps plan beforehand, the demand, supply, and storage of water. This irrigation depends on the weather predictions, soil conditions and crop type [2]. Smart irrigation has advanced significantly because of the merger of technologies such as Artificial Intelligence (AI), Machine Learning (ML) and big data, with IoT. It has been realized that smart irrigation is essential in agriculture 4.0 and is crucial for increasing the global food production efficiency. Agriculture 4.0 aims to not just grow more food with minimum resources, but also improve efficiency of the overall ecosystem, by incorporating powerful data analysis, intelligent automation, and decisionmaking [3], [4] and [5].