"Semantic-Aware Sensor Anomaly Detection Dataset"

"This dataset contains multivariate time-series measurements collected from an embedded sensing platform designed for research in anomaly detection and fault diagnosis in sensor-based control systems. The platform is built using an ESP32 microcontroller interfaced with heterogeneous sensors, including temperature, humidity, and optical sensing (photoresistor), along with controllable RGB LED actuation signals.Data were collected under both normal operating conditions and controlled fault injection scenarios to simulate realistic sensor and system anomalies such as signal noise, saturation, and actuation\u2013sensor inconsistencies. Each record includes synchronized sensor measurements, actuator states, timestamps, and ground-truth fault labels, enabling supervised evaluation of anomaly detection and root-cause analysis algorithms.The dataset is intended to support research in industrial IoT monitoring, cyber\u2013physical systems, semantic-aware sensor analytics, and machine learning\u2013based diagnostic modeling. The structured nature of the dataset allows evaluation of both time-series-only and context-aware fault detection approaches. This dataset is suitable for benchmarking anomaly detection models, developing interpretable diagnostic algorithms, and studying sensor\u2013actuator interaction behavior in embedded sensing environments."