Cognitive sensor networks for structure defect monitoring and classification using guided wave signals

This paper develops a framework of a cognitive sensor networks system for structure defect monitoring and classification using guided wave signals. Guided ultrasonic waves that can propagate long distances along civil structures have been widely studied for inspection and detection of structure damage. Smart ultrasonic sensors arranged as a spatially distributed cognitive sensor networks system can transmit and receive ultrasonic guided waves to interrogate structure defects such as cracks and corrosion. A distinguishing characteristic of the cognitive sensor networks system is that it adaptively probes and learns about the environment, which enables constant optimization in response to its changing understanding of the defect response. In this paper, we develop a sequential multiple hypothesis testing scheme combined with adaptive waveform transmission for defect monitoring and classification. The performance is verified using numerical simulations of guided elastic wave propagation on a pipe model and by Monte Carlo simulations for computing the probability of correct classification.