Identification of vibration modes from sub-microstrain fiber-optic Bragg Grating data using an improved wavelength detection algorithm

{Fibre-optic Bragg Grating (FBG) strain sensors hold a great potential for vibration monitoring of civil structures because of their exceptional stability and accuracy; however, the accurate measurement of the very small strains levels occurring during ambient, or operational, excitation has been so far problematic. There are two ways to improve the measurement resolution: employing a strain-enhancing sensor package, and applying an improved wavelength detection algorithm. In this work, the potential of an improved wavelength detection algorithm for the identification of modal characteristics based on sub-microstrain data is investigated. The strategy is illustrated for a steel beam to which a chain of multiplexed FBG sensors has been attached at the top side of the beam. The raw FBG data are processed into strain values with an algorithm that is based on detecting the peak shifts in the wavelength spectrum by correlation analysis rather than by simply tracking the peak values. Subsequently, the strain sequences are used for identification of modal characteristics of the beam (natural frequencies and strain mode shapes) with ``Covariance driven Stochastic Subspace Identification (SSI/cov){''}. Computational results of a Finite Element Model are used to validate the experimental results.}