Wavelet-Based Vibration Sensor Data Compression Technique for Civil Infrastructure Condition Monitoring

Abstract

Civil infrastructure condition monitoring generates large volumes of sensor data. Huge data size impedes fast and reliable distribution of sensor data, especially for wireless sensor networks. Vibration sensor data plays an important role in many useful structural health monitoring methods. This paper presents a vibration sensor data compression technique based on the lifting scheme wavelet transform (LSWT). Real sensor data from a nine-story building as well as a steel truss bridge was used to examine the compression performance of the LSWT-based vibration sensor data compression technique. It is found that the LSWT-based vibration sensor data compression technique can achieve a very high compression ratio while retaining the basic waveform properties of original sensor data. Additionally, LSWT has a feature that supports progressive compression and thus allows for multiresolution data transmission and retrieval.