Offline Synchronization of Signals From Multiple Wireless Sensors

This study addresses the critical challenges of time synchronization in wearable sensor networks, focusing on electrocardiogram (ECG) and inertial measurement unit (IMU) monitoring applications. In the era of continuous physiological and biomechanical monitoring, accurate time synchronization of sensor data is critical. This article investigates the effectiveness of an offline synchronization method chosen for its flexibility and precision in addressing time-related anomalies in environments where real-time processing of the gathered data is not required. The synchronization method works independently for each node without exchanging time-sync packets among nodes, only among nodes and a central device. We present a synchronization approach that has been designed to deal with variable sampling frequency, random transmission delay, and packet loss. We demonstrate the efficiency of the approach with two different example applications: long-term ECG monitoring and short-term IMU-based gait analysis. The example applications use different strategies for storing the sampled data and for exchanging time-sync packets. Our results show that the proposed synchronization method is robust and accurate. We identify the limit for accuracy to be in the communication software of the controller device and sensor nodes. This study contributes to the field of wearable sensor networks by presenting a comprehensive synchronization method.