LMR: A software for modeling and reverse optimization of straight-core lossy mode resonance (LMR) based sensors

Abstract

Lossy mode resonance (LMR) is a physical phenomenon recently exploited for fiber optic sensing. LMR-based devices are widely used for detecting refractive index variation, humidity, pH (acidity or basicity of an aqueous solution), chemical or biological species, gases, or even voltage. Two main types of geometries can be distinguished: prism-based and waveguide-based. In both cases, however, the manufacture and prototype development of the sensor require exact and expensive technology. Therefore, reliable and fast modeling of these devices is desired to reduce the costs of their investigation, design, and production. LMR is an R-toolbox for simulating LMR sensors of straight-core geometry (the most common waveguide-based type). The mathematical model is based mainly on geometrical optics, while the equations of classical electromagnetism and the transfer matrix method were used to determine the reflection coefficients from stratified media. This combined approach is why we call this model hybrid. It may seem simple, but it is perfectly sufficient to predict the most important parameters of the sensor. The results of calculations were experimentally verified in many works. The package allows one to load up to 8 layers in the sensing region and provides access to the material database. The researcher can use the dependencies contained therein to describe the complex refractive index as a function of the wavelength of a given material or use his own data. Another essential feature of the package is reverse optimization capability. The applied genetic algorithm (GA) allows the design of a sensor with a given transmission spectral characteristics. Using the program is facilitated by an intuitive graphical interface. The final result is the transmission of the optical fiber in decibels versus the wavelength.