Self-Tuning Model for Energy-Context Optimization in Perpetual Sensor Nodes Within IoT-Integrated Hydroponic Systems

Abstract

Hydroponic farming is a promising alternative to soil-based farming. However, it requires a precise control of the growth environment, which is hard to achieve with energy-constrained embedded systems. This letter presents an energy optimization technique for the continuous operation of energy harvesting-based hydroponics sensor nodes. The proposed technique is based on the self-tuning model, that dynamically adjust the duty cycle of the node, ensuring the autonomous operation of the Internet of Things system. The model can be programmed in a low-power microcontroller, allowing the decision-making process to reside entirely on the sensor node. Experimental results show that in the same time period, the self-tuning model allows $3.5\times $ more data transmissions than a uniform 5-min duty cycle, while ensuring a minimum voltage level in the storage device. This balance allows the stored energy to be enough for continuous monitoring, providing a clean and cost-effective alternative to perpetually power the hydroponic system.