Internet of Things (IoT) Sensor Networks for Real-Time Assessment of Animal Health and Environmental Conditions

Abstract

As the need for constant animal health and environmental conditions monitoring has grown, the traditional field inspections have proven to be time-consuming, labor intensive and cannot give immediate warnings. The goal of this study is to design an Internet of Things (IoT)-based sensor network for real-time monitoring of animal health and environmental conditions around them for precision livestock management and wildlife conservation. These goals are to continuously monitor physiological parameters (body temperature, heart rate and activity), measure environmental parameters (temperature, humidity, air quality and noise) and detect abnormal conditions intelligently based on analytics. The proposed approach combines the use of wearable IoT sensors, LoRa/Wi-Fi communication, edge computing, cloud storage, and an anomaly detection model based on the Random Forest algorithm, and a web-based monitoring dashboard. The results are obtained by experimental evaluation of 12,480 sensor readings from 85 animals over a period of 90 days, and proves 94.82% accuracy in the health status classification, 93.76% accuracy in the prediction of environmental conditions, 92.91% accuracy in anomaly detection, 93.48% accuracy in recall, 93.19% accuracy in F1-score, 96.24% of sensor data is sent successfully, and 91.37% of network is up and running with an average alert response time of 2.8 s, while consuming 18.6% less energy than conventional cloud-only architecture. The novelty of this work is the combination of edge-assisted IoT sensing with intelligent anomaly detection that can be used to monitor the animal health and environment simultaneously. The framework, as proposed, is reasonably energy efficient, scalable, and reliable that will enable early detection of disease, quick environmental risk assessment and data-informed decision-making to ensure sustainable livestock and wildlife management.