Development of a Wireless Sensor Network and IoT-based Smart Irrigation System

Abstract

This study proposes a smart irrigation system (SIS) using the drip method, which was designed and implemented using wireless sensor networks and an open-source Internet of Things (IoT) cloud computing platform (“Thingspeak.com”) for data collection, storing, data analytics, and visualization. The methodology incorporates the integration of hardware and software components to make irrigation decisions based on web resources like the weather forecast from “weather.com” and sensor values from soil samples. The data collected are then analyzed at the edge server and updated every 15 minutes. Based on the threshold value, the system starts pumping water or stops the irrigation process depending on the irrigation schedule. A web application was developed to display the result so that we could monitor and control the system using an android application edge or a web browser. Based on the data recorded and measured, the data are in comma-separated values (CSV) format and contain 143731 entries with 10 columns. The sample size used contains 5722 rows and 6 columns from the result of our machine learning algorithms using Microsoft Excel and Jupyter Notebook to process and evaluate the performance of the drip SIS, considering the soil moisture, soil temperature, sunlight, rain, and pump. The results confirm the threshold metrics classification evaluation, and some of the metrics computed from our confusion matrix are shown in the classification summary results, showing the accuracy to be 89%, the misclassification rate (error rate) is equal to 10%, the sensitivity is equal to 79%, the specificity is equal to 93%, and the precision of the model is 81%. The evaluation, when compared to K-nearest neighbours using K = 6 and K = 1, shows the prediction accuracy to be 97% and 98%. The results indicate the system is highly efficient and reliable in performing irrigation and managing water resources and can be adopted in rural areas to boost agricultural productivity.

An adaptive neuro-fuzzy inference system-based irrigation sprinkler system for dry season farming

“…Another interesting finding is that the system can adjust in real time to changes in the weather. Changes in the weather may have a significant impact on agricultural production, making it essential to have a system that can dynamically modify water delivery depending on real-time circumstances [23]- [34]. The system's intelligence and adaptability are on full display in its abilities to optimize water distribution in the face of adverse weather.…”

Developing a Hybrid Irrigation System for Smart Agriculture Using IoT Sensors and Machine Learning in Sri Ganganagar, Rajasthan

“…The system achieves 89% accuracy, a 10% misclassification rate, 79% sensitivity, 93% specificity, and 81% precision in threshold metrics classification evaluation. The system's 97% and 98% prediction accuracy indicate reliable and efficient irrigation and water resource management, potentially boosting agricultural production in rural areas [36] is used to measure soil moisture content in the field. The required soil moisture threshold value is described, and if soil moisture exceeds or decreases from the threshold value, the soil moisture sensor will send the current soil moisture value to the server through the node MCU.…”

Ontology-Based Smart Irrigation System: Enhancing Agricultural Water Management

“…The irrigation schedule is used to determine whether or not to begin pumping water based on the threshold value. A web application was developed to monitor and control the system based on the data collected (Ndunagu et al, 2022). All forms of life, from plants to animals to humans, require water in order to survive.…”

“…The process of watering a surface specifically so that vegetation might flourish there is known as irrigation. A body of water is guided by canals, ditches, pipelines, or even just the course of the water itself (Ndunagu, et al, 2022). In arid regions, proper water management is crucial.…”