Sensor Network System with Dynamic Service Deployment Depending on Network Condition for Smart Aquaponics

Abstract

There is an urgent need to automate and improve the efficiency of agricultural work, and IoT systems are attracting attention to increase the efficiency of agricultural operations. This study focuses on Aquaponics, an efficient agricultural method that combines hydroponics and aquaculture. Aquaponics operations need to not only manage the environment for plants, bacteria, and fish but also ensure that the plants and fish are growing properly. Therefore, we propose sensor network system that aims to improve the efficiency of aquaponics operations by analyzing information on the state of the growing environment obtained from various sensors, as well as the growth rate of plants and fish. Furthermore, to reduce the workload of administrators for device management, we propose an efficient management system with a container orchestration platform that dynamically deploys applications to nodes installed in the field. The system adopts a highly available architecture for farms where the condition of the network communication environment fluctuates. The proposed architecture is equipped with dynamic service deployment that automatically switches between edge and cloud computing based on the network conditions. Through performance evaluation, we confirm that containers equipped with the proposed analysis method can be automatically deployed and operated on edge devices using the container orchestration platform.