An Automated Approach to Collecting and Labeling Time Series Data for Event Detection Using Elastic Node Hardware
Tianheng Ling, Islam Mansour, Chao Qian, Gregor Schiele
TL;DR
This work addresses the challenge of efficiently collecting and labeling time-series sensor data for event detection on IoT devices. It introduces the Elastic Node Sensor Logger, a hardware-software platform that performs lightweight, autonomous labeling on-device, reducing data transmission and external dependencies. Empirical results from three event types show high-quality labeled data, with CNN-based time-series classification achieving up to 91.67% accuracy under 4-fold cross-validation. The approach demonstrates the practical impact of edge-enabled data labeling for scalable, reliable IoT sensing, and outlines future work to expand sensor types for broader applicability.
Abstract
Recent advancements in IoT technologies have underscored the importance of using sensor data to understand environmental contexts effectively. This paper introduces a novel embedded system designed to autonomously label sensor data directly on IoT devices, thereby enhancing the efficiency of data collection methods. We present an integrated hardware and software solution equipped with specialized labeling sensors that streamline the capture and labeling of diverse types of sensor data. By implementing local processing with lightweight labeling methods, our system minimizes the need for extensive data transmission and reduces dependence on external resources. Experimental validation with collected data and a Convolutional Neural Network model achieved a high classification accuracy of up to 91.67%, as confirmed through 4-fold cross-validation. These results demonstrate the system's robust capability to collect audio and vibration data with correct labels.