Adaptation of state-of-the-art neural network architectures to interference fringe reduction in absorption spectroscopy

Abstract

State-of-the-art neural network architectures in image classification and natural language processing were applied to interference fringe reduction in absorption spectroscopy by interpreting the data structure accordingly. A model was designed for temporal interpolation of background spectra and a different model was created for gas concentration fitting. The networks were trained on experimental data provided by a wavelength modulation spectroscopy instrument and the best performing architectures were analyzed further to evaluate generalization performance, robustness and transferability. A BERT-styled fitter achieved the best performance on the validation set and reduced the mean squared error of fitted amplitude by 99.5%. However, analysis of the de-noising behavior showed large biases. A U-Net styled convolutional neural network reduced the mean squared error of the interpolation by 93.2%. Evaluation on a test set provided evidence that the combination of model interpolation and linear fitting was robust and the detection limit was improved by 52.4%. Transferring the trained interpolator model to a different spectrometer setup showed no chaotic out-of-distribution effects. Additional fine-tuning further increased the performance. Neural network architectures cannot be generally applied to all absorption spectroscopy tasks. However, given the right task and the data representation, robust performance increase is achievable.