Advanced methods for fiber-optic sensor signal processing
Radek Martínek, Radana Vilimkova Kahankova, Jan Nedoma, Marcel Fajkus, Martin Novák, Jan Jargus · 2018
This article is focused on the advanced signal processing methods for third-generation sensors requirements. These sensors are based on the influence of a non-electric quantities on a light beam. This generation of sensors, also known as fiber optic sensors, is based on the principles of optoelectronics and integrated optics. These sensors are used in a variety of real-world applications such as biomedical engineering, industry 4.0, transportation, etc. In real-world applications, the signals sensed by these sensors are distorted by a variety of interference due to its sensitivity. We often encounter the problem that the useful information and the interference overlap in the spectral domain, therefore we cannot use conventional frequency selective filters. This article focuses on the implementation of adaptive filtering, Principle Component Analysis and Independent Component Analysis to reduce the interference in various application areas. The methods were tested on real data. This paper offers the comparison of the tested methods in different application areas.