Wireless sensor networks for high fidelity sampling

Our hypothesis was that Wireless Sensor Networks (WSN) could be used for High Fidelity Sampling (HFS). WSN is constrained by limited resources, and HFS requires data of high quality, like high-accuracy high frequency in a large scale. This dissertation explains how we achieved our goal. A Wireless Sensor Network (WSN) for Structural Health Monitoring (SHM) is designed, implemented, deployed and tested on the 4200ft long main span and the south tower of the Golden Gate Bridge (GGB). Ambient structural vibrations are reliably measured at a low cost and without interfering with the operation of the bridge. Requirements that SHM imposes on WSN are identified and new solutions to meet these requirements are proposed and implemented. For example, diverse design options for a reliable data transfer are analyzed and an initial solution for a reliable data collection, Straw, is proposed. In the GGB deployment, 64 nodes are distributed over the main span and the tower, collecting ambient vibrations synchronously at 1kHz rate, with less than 10µs jitter, and with an accuracy of 30µG. The sampled data is collected reliably over a 46-hop network, with a bandwidth of 441B/s at the 46th hop. The deployment is the largest WSN for SHM. Data from the deployment is analyzed. The collected vibration data agrees with theoretical models and previous studies of the bridge. Interesting behaviors are observed, and pitfalls and lessons are discussed. Especially, to overcome pitfalls of reliable data collection from the deployment, an improvement, Flush, is proposed. The Flush protocol provides an end-to-end reliability, minimizes transfer time, and adapts to time-varying network conditions. It achieves these properties using end-to-end acknowledgments, implicit snooping of control information, and a rate-control algorithm that operates at each hop along a flow. Using several real network topologies, we show that Flush closely tracks or exceeds the maximum goodput achievable by a hand-tuned, fixed-rate for each hop over a wide range of path lengths.