Passive impacts localization based on dispersion compensation and cross-correlated signals wavelet analysis

Alessandro Perelli, Luca De Marchi, Alessandro Marzani, Nicolò Speciale

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

A method for impact location in plate-like structures with a passive sensors network is proposed. The approach is based on guided waves dispersion compensation joint with a distance-frequency analysis obtained through a wavelet basis of cross-correlated signals. In passive monitoring techniques the knowledge of the time of impact is not given; despite this limit, the proposed dispersion compensation procedure is useful as it removes in the group delay of the acquired signals the dependence on the travelled distance. By cross-correlating the signals related to the same event acquired by different sensors, the time difference of arrival is estimated. To reduce the interference due to the edge reflections and the own plate echoes, the cross-correlating signal is decomposed by a suitable orthogonal basis and the magnitude of the Continous Wavelet Transform is used to obtain the difference in travelled distances and to locate the wave source via hyperbolic positioning. The proposed procedure is tested with a passive network of three/four piezo-sensors located symmetrically and asymmetrically with respect to the plate edges. The experimentally results are close to those theoretically predicted by Cramèr-Rao bound.
Original languageEnglish
Title of host publicationAIP Conference Proceedings
Pages431-434
Number of pages4
Volume1433
DOIs
Publication statusPublished - 24 May 2012
EventInternational Congress on Ultrasonics 2011 - Gdańsk, Poland
Duration: 5 Sep 20118 Sep 2011

Conference

ConferenceInternational Congress on Ultrasonics 2011
Abbreviated titleICU 2011
Country/TerritoryPoland
CityGdańsk
Period5/09/118/09/11

Fingerprint

Dive into the research topics of 'Passive impacts localization based on dispersion compensation and cross-correlated signals wavelet analysis'. Together they form a unique fingerprint.

Cite this