A computational framework for modeling complex sensor network data using graph signal processing and graph neural networks in structural health monitoring

Complex networks lend themselves for the modeling of multidimensional data, such as relational and/or temporal data. In particular, when such complex data and their inherent relationships need to be formalized, complex network modeling and its resulting graph representations enable a wide range of powerful options. In this paper, we target this—connected to specific machine learning approaches on graphs for Structural Health Monitoring (SHM) from an analysis and predictive (maintenance) perspective. Specifically, we present a framework based on Complex Network Modeling, integrating Graph Signal Processing (GSP) and Graph Neural Network (GNN) approaches. We demonstrate this framework in our targeted application domain of SHM. In particular, we focus on a prominent real-world SHM use case, i. e., modeling and analyzing sensor data (strain, vibration) of a large bridge in the Netherlands. In our experiments, we show that GSP enables the identification of the most important sensors, for which we investigate a set of search and optimization approaches. Furthermore, GSP enables the detection of specific graph signal patterns (i. e., mode shapes), capturing physical functional properties of the sensors in the applied complex network. In addition, we show the efficacy of applying GNNs for strain prediction utilizing this kind of sensor data.

Data Reconstruction Coverage Based on Graph Signal Processing for Wireless Sensor Networks

Sensing coverage is a crucial metric for the quality of service of Wireless Sensor Networks (WSNs). Coverage models have a great impact on sensing coverage of WSNs. However, existing coverage models are simple but inefficient, like the most frequently used disk coverage model, in which a covered point is within the fixed sensing radius of at least one sensor node. Thus, how to develop an efficient coverage model is an essential problem. To this end, in this letter, we propose a novel coverage model without the limitation of the sensor’s sensing radius, namely, Data Reconstruction Coverage (DRC). Based on the theory of graph signal processing, the model can jointly reconstruct missing data at unsampled points (which are not covered by any sensors) by using our proposed centralized data reconstruction coverage algorithm which fully exploits the smoothness of temporal difference signals and the graph Laplacian matrix, without increasing the number of sensors. Simulation results based on real-world datasets show that the proposed DRC model has better coverage performance of WSNs compared with the disk coverage model and confident information coverage model typically used in WSNs.

Data Reconstruction Coverage Based on Graph Signal Processing for Wireless Sensor Networks

Sensing coverage is a crucial metric for the quality of service of Wireless Sensor Networks (WSNs). Coverage models have a great impact on sensing coverage of WSNs. However, existing coverage models are simple but inefficient, like the most frequently used disk coverage model, in which a covered point is within the fixed sensing radius of at least one sensor node. Thus, how to develop an efficient coverage model is an essential problem. To this end, in this letter, we propose a novel coverage model without the limitation of the sensor’s sensing radius, namely, Data Reconstruction Coverage (DRC). Based on the theory of graph signal processing, the model can jointly reconstruct missing data at unsampled points (which are not covered by any sensors) by using our proposed centralized data reconstruction coverage algorithm which fully exploits the smoothness of temporal difference signals and the graph Laplacian matrix, without increasing the number of sensors. Simulation results based on real-world datasets show that the proposed DRC model has better coverage performance of WSNs compared with the disk coverage model and confident information coverage model typically used in WSNs.