Analysis of multiparameter optical sensor data with composite filtering algorithms

This paper presents a novel approach to the analysis of detection signals from optical sensors. The proposed technique is based on the pattern recognition algorithms that belong to the family of composite linear filters and is compared against a MAP-estimate approach. Our study is focused on the information recovery from the reflectivity spectrum of a Bragg sensor measured with an optical spectrum analyzer. However we note that the same scheme would be equally useful for the extraction of information obtained with a polarimetric fibre sensor or an optical chemical sensor, where in general the detection signal is multidimensional and includes data related to several measured parameters.