"IoT Sensor Dataset for Safety Monitoring: Temperature, Pressure, and Jerk Events"

"This dataset presents a comprehensive collection of time-series sensor data acquired from an ESP32-based IoT safety monitoring system deployed in a real-world indoor environment. The system continuously records synchronized measurements of temperature (\u00b0C), atmospheric pressure (hPa), and jerk (m\/s\u00b2) at one-second intervals, enabling fine-grained temporal analysis of environmental and motion-related dynamics.The jerk signal, representing the rate of change of acceleration, is derived from raw accelerometer readings and serves as a critical feature for identifying sudden movements, impacts, and transitions between activity states such as walking, running, stationary conditions, and free-fall. By combining environmental parameters with motion-derived metrics, the dataset offers a multi-modal perspective that enhances the detection and interpretation of both routine patterns and anomalous events.Data collection was conducted over several consecutive days under natural operating conditions, capturing a diverse range of real-world scenarios, including normal daily activities as well as spontaneous abnormal events such as abrupt impacts and rapid motion changes. This ensures the dataset reflects realistic noise, variability, and sensor behavior typically encountered in practical IoT deployments.The dataset is well-suited for a wide range of research and development applications, including anomaly detection in streaming sensor data, activity recognition and classification, safety monitoring in industrial and healthcare environments, and the development of machine learning models for predictive analytics. Additionally, it can serve as a benchmark dataset for evaluating time-series analysis techniques, edge computing algorithms, and real-time event detection systems in resource-constrained embedded environments.Overall, this dataset provides a reliable and practical foundation for advancing research in IoT-based sensing, intelligent safety systems, and data-driven decision-making in cyber-physical systems."