Comparison of different peak detection algorithms with regards to spectrometic fiber Bragg grating interrogation systems

The resolution of many optical measurement systems employing ID or 2D array sensors (e.g. CCD-cameras) is limited by the pixel-resolution of the detector. Subpixel algorithms allow to exceed this limit. This paper aims at fiber Bragg grating interrogators, which acquire the sensor signal spectrometrically. The accuracy of measurement of these systems is strongly depending on the used algorithms. Five different algorithms for peak detection are described and compared in theory and experiment. Most of these algorithms can be found both in literature and in different software-libraries (e.g. MATLAB, LabVIEW...) To compare the different algorithms, a single fiber Bragg grating sensor was used to acquire data with a spectrometric measurement system consisting of a lD-CCD-line-array, a superluminescent light source and a reflection grating. The same data was fed the five algorithms, such that the output signals are comparable. To characterize the peak-search algorithms the standard deviation of the output data has been computed for different frame rates and variable wide regions of interest around a FBG-peak. As a result we can provida a concise recommendation which of the analyzed algorithms is suitable for an application in spectrometric fiber Bragg grating interrogators.