Design and implementation of distributed algorithms for WSN-based structural health monitoring

Wireless Sensor Networks (WSNs) have found many applications in a variety of ubiquitous computing environments. One of the key applications is Structural Health Monitoring (SHM) which seeks to monitor the functionality and detect the possible damages of large civil engineering structures, by measuring structural responses to ambient or forced excitation. WSNs are a promising new technology for SHM, but the contradicting considerations on the unique characteristics of WSNs and the inherent requirements of SHM pose challenges for practical implementation. Distributed processing in WSNs can help reduce energy consumption, make efficient use of network bandwidth and improve system response time. In this paper, we analyse the design issues of WSN-based SHM, and propose distributed processing architecture with algorithms. We also describe our system prototype built using Crossbow MICAz nodes and custom-made TFAD-901 super nodes. The performance of the proposed algorithms has been evaluated using various approaches. The results of the emulation study using the prototype system demonstrate the effectiveness of the proposed distributed processing approach. The results of laboratory testing and in-field experiments show the feasibility of practical deployment. Finally, the results of simulations enable us to gain insight of the distributed algorithms in large-scale systems.