Graph Signal Processing on Complex Networks for Structural Health Monitoring

Abstract

In this work, we demonstrate the application of a framework targeting Complex Networks and Graph Signal Processing (GSP) for Structural Health Monitoring (SHM). By modeling and analyzing a large bridge equipped with strain and vibration sensors, we show that GSP is capable of selecting the most important sensors, investigating different optimization techniques for selection. Furthermore, GSP enables the detection of graph signal patterns (mode shapes), grasping the physical function of the sensors in the network. Our results indicate the efficacy of GSP on complex sensor data modeled in complex networks. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.