Monitoring the dynamic performance of structures usually requires a deployment of several weeks or months to collect a data set allowing a reliable assessment. Since for this type of medium term deployments installation costs are a key factor, wireless monitoring with its fast deployment has an advantage over wired monitoring systems. This paper describes a deployment of a wireless monitoring system on a timber footbridge. The goal of the monitoring was to provide information about vibration amplitudes during operation and to track the changes of relevant natural frequencies with temperature. The monitoring data quality matched the requirements of a dynamic performance assessing process. The wireless monitoring system worked for about one year with very high reliability. Battery replacement was necessary only once after 6 months of operation. The deployment demonstrated that a wireless sensor network is a technically feasible and economically effective mean to monitor the dynamic performance of a structure.

In the last decade, wireless sensor networks have become an intensively investigated tool for monitoring applications. Many field experiments with short-term deployments demonstrated their advantages. Very little, however, is known about the performance of wireless sensor network in mid- and long-term deployments. This paper reports about the experiences in developing, deploying and maintaining a wireless monitoring system on a cable stay bridge.

Structural health monitoring with wireless sensor networks has received much attention in recent years due to the ease of sensor installation and low deployment and maintenance costs. However, sensor network technology needs to solve numerous challenges in order to substitute conventional systems: large amounts of data, remote configuration of measurement parameters, on-site calibration of sensors and robust networking functionality for long-term deployments. We present a structural health monitoring network that addresses these challenges and is used in several deployments for monitoring of bridges and buildings. Our system supports a diverse set of sensors, a library of highly optimized processing algorithms and a lightweight solution to support a wide range of network runtime configurations. This allows flexible partitioning of the application between the sensor network and the backend software. We present an analysis of this partitioning and evaluate the performance of our system in three experimental network deployments on civil structures.

In the last decade, wireless sensor networks have emerged as a promising technology that could accelerate progress in the field of structural monitoring. The main advantages of wireless sensor networks compared to conventional monitoring technologies are fast deployment, small interference with the surroundings, self-organization, flexibility and scalability. These features could enable mass application of monitoring systems, even on smaller structures. However, since wireless sensor network nodes are battery powered and data communication is the most energy consuming task, transferring all the acquired raw data through the network would dramatically limit system lifetime. Hence, data reduction has to be achieved at the node level in order to meet the system lifetime requirements of real life applications. The objective of this paper is to discuss some general aspects of data processing and management in monitoring systems based on wireless sensor networks, to present a prototype monitoring system for civil engineering structures, and to illustrate long-term field test results.

In the last years, wireless sensor networks have emerged as a promising technology that is inducing a deep innovation in the field of structural monitoring. The main advantages of wireless sensor networks are fast deployment, little interference and selforganization. However, since wireless sensor nodes are battery powered, in long term monitoring applications the power management influences significantly the operation of a wireless sensor network. In data intensive applications, e.g. vibration based monitoring, low power hardware, duty cycle operation and efficient communication policies are not sufficient for achieving a sustainable system lifetime. Since data communication is the most energy consuming task, long system lifetimes can only be achieved by a significant data reduction in the nodes. This data reduction is a challenging task, since it has to be performed with very limited computational and memory resources and in competition with tasks providing the basic network functionality. The objective of the paper is to provide a brief overview of the wireless sensor network technology and to present our experience over the past three years with data intensive structural monitoring using wireless sensor networks. Deployments on two bridges are illustrated and specific aspects of sensing, data quality, stability, availability, and system lifetime are analyzed. 1 WIRELESS SENSOR NETWORKS

This paper presents a monitoring application with a wireless sensor network that was performed on a 95 years old riveted steel railway bridge. In order to perform an accurate assessment, strains we ...

Wireless sensor networks (WSNs) are a promising technology that could induce a significant innovation in the field of structural monitoring. The main advantages of WSNs are fast deployment, little interference and self-organisation. However, since WSN are battery powered, the power management of the sensor nodes significantly influences the operation method and the overall data management process. Since data communication is the most energy-consuming task, a significant data reduction has to be attained in the sensor nodes to achieve system lifetimes that are useful for real life applications. This paper discusses several basic aspects of data processing and data management for long-term monitoring with WSNs. It presents a specific monitoring system and illustrates a long-term field test performed with this system on a bridge. The test results demonstrate that in-network data reduction is a very promising but challenging approach, since it has to be implemented with very limited computational and memory resources.

The support for the actual deployment of wireless sensor networks is, notwithstanding an increased interest and work in this field, still an underdeveloped area of research. We discuss the use of two simple objects built from household materials -- a cantenna and a tinfoil cylinder -- to increase the directivity of an antenna of a standard mote. This MoteFinder can be used in a variety of applications including node localization and as a tool for selective communication with groups of nodes. We show in our evaluation that both devices provide a good sense of direction in indoor and outdoor scenarios and provide a foundation for future research.

Microelectromechanical system (MEMS) sensors are small, generally low power, highly integrated, and, usually, very cheap. These qualities enable the deployment of structural health monitoring (SHM) systems with a large number of sensors, partly integrated into the structure, at affordable costs. MEMS sensors are often used in wireless sensor networks (WSNs), a monitoring technology that heavily bases its features and performance on low‐power sensors. This article briefly describes a MEMS‐based wireless sensor network that is designed for long‐term structural health monitoring applications. Since WSN nodes are battery powered, in long‐term monitoring applications the power management of the network significantly influences the overall data handling processes. In WSN, data communication is the most energy‐consuming task, hence, transferring all the acquired raw data through the network would dramatically reduce the lifetime of the system. A significant data reduction, which has to be achieved on the nodes, is a challenging task, since it has to be performed with very limited computing power and memory resources, and in competition with other tasks like data communication, self‐organization, and time synchronization. These aspects are illustrated with tests on a cable stay bridge.

AbstractA wireless sensor network is a network made up of many tiny intercommunicating computers equipped with one or several sensors. Each tiny computer represents a node of the network. The nodes are self‐contained units typically consisting of a power supply with limited capacity, a radio transceiver, a microcontroller, and one or more sensors. This article gives an introduction to wireless sensor networks for structural health monitoring, shows the general architecture of sensor nodes, and overviews current hardware and software platforms. Furthermore, it shows the characteristics and limits of such monitoring systems and gives advice to choose a suitable platform to an application. Because nodes are powered by an autonomous source, energy‐related aspects, energy storage, and scavenging are presented as well.

The objective of this paper is to present a prototype wireless sensor network for monitoring of civil engineering structures. The network consists of several remote sensor nodes distributed over a structure, representing the data sources and a base station (data logging and configuration unit), representing the data sink in the network. Each node is equipped with several sensors, a data acquisition and signal processing unit and a radio transceiver. The acquired data is pre-processed on the sensor node before it is sent to the base station. The aggregated data can be remotely accessed over the Internet. Software tools permit to administrate the network and re-schedule measurement tasks remotely. The acquired data is stored in a relational database that can be accessed and updated by different data query and visualization tools. After preliminary laboratory experiments a sensor network has been deployed on a cable stayed road bridge to evaluate the performance of the system. The deployed system monitors the tension force of the stays as well as temperature and humidity. The force is monitored by tracking natural frequencies estimated from measured ambient cable vibrations.

The objective of this paper is to present a prototype monitoring system for civil structures based on a wireless sensor network. The sensor network is composed of a root node, representing the data sink, logging and configuration unit in the network and several sensor nodes (motes) distributed over a structure and representing the data sources. Each mote is equipped with sensors, a digital signal processing unit, a radio transceiver and a power supply. All motes together form a multi-hop communication network. That is, each mote figures as a data source and a relaying station, forwarding data received from adjacent motes towards the base station, i.e. the data sink. The data received at the base station is stored in a database where it can be accessed for further post-processing. The accessibility is provided by various software tools. These tools allow a user or system operator to monitor the condition of the structure and enable the operator to configure and administrate the monitoring system remotely. This paper additionally presents laboratory tests performed with the prototype monitoring system on the scaled cable stayed pedestrian bridge at Empa. The prototype system monitors the stay cable forces. The force is calculated via natural frequency estimation of cable accelerations. This computation is executed on the motes. Only the cable forces are transmitted to the root node. The root node is connected to a terminal which collects and visualizes the data received from the motes and performs a pre-diagnosis for identifying abnormal states.

A cable stayed footbridge with a glass fiber reinforced polymer girder has been erected in the Swiss Federal Laboratories for Materials Testing and Research (Empa). The main purpose of this project is to close the gap between complex real-world applications in civil engineering and simplified laboratory experiments. The bridge is a research platform in the fields of active, semiactive and passive vibration mitigation, structural health monitoring, integrated distributed smart sensing and advanced structural materials. The bridge will be equipped with several interconnected sensors and actuators. This paper discusses different aspects of such instrumentation and communication techniques.

We present an image-space technique, which can detect intersections and self-intersections among multiple moving and deforming objects. No preprocessing is needed and the shape of the objects are unconstrained and can be an arbitrarily polygon soup. Compared to other intersection detection algorithms running on graphics hardware the algorithm only make modest use of bandwidth between the CPU and GPU because no buffer readbacks are necessary.

We present an image-space technique, which can detect intersections and self-intersections among multiple moving and deforming objects. No preprocessing is needed and the shape of the objects are unconstrained and can be an arbitrarily polygon soup. Compared to other intersection detection algorithms running on graphics hardware the algorithm only make modest use of bandwidth between the CPU and GPU because no buffer readbacks are necessary.

This paper presents a prototype monitoring system for civil structures based on a wireless sensor network. The sensor network is composed of a root node, representing the data sink, logging and configuration unit and many sensor nodes (motes) distributed over a structure. Each mote is equipped with sensors, a digital signal processing unit, a radio transceiver and a power supply. All motes together form a multi-hop communication network. That is, each mote figures as a data source and a relaying station, forwarding data received from adjacent motes. The routing of the data packets is optimized for maximizing the network’s lifetime. Another feature of the network is its capability of self-organization. If a single mote or parts of the network fail, new paths to the root node are discovered and established. Low power consumption is a key issue in wireless sensor networks because the motes have to operate from batteries for a reasonably long time. Since data communication dissipates remarkable more energy than data processing, it is essential to execute suitable data reduction algorithms allowing for a significant reduction of data items which have to be transmitted to the root node. The base station of the sensor network is accessible over the internet using wired or wireless connections. A locally or remotely connected terminal visualizes the data and performs a prediagnostic for identifying abnormal states. Moreover it allows for administration of the nodes and configuration of measurement setup. Additionally the collected data are stored in a database. This paper presents laboratory tests performed with the prototype monitoring system on the scaled cable stayed pedestrian bridge at Empa. The prototype system monitors the stay cable forces. The force is calculated via natural frequency estimation of cable accelerations. This computation is executed on the motes. Only the cable forces are transmitted to the root node. _____________ Reinhard Bischoff, Jonas Meyer, Glauco Feltrin, Structural Engineering Research Laboratory, Swiss Federal Laboratories for Materials Testing and Research, Ueberlandstrasse 129, 8600 Duebendorf, Switzerland, reinhard[dot]bischoff[at]empa[dot]ch Olga Saukh, University of Stuttgart, Universitaetsstrase 38, 70569 Stuttgart, Germany

This report describes procedures and algorithms for data analysis and reduction methodologies regarding vibration analysis and acoustic emission techniques (AET) utilized for a prototype bridge mon ...