Generalized Radar-Based Detection of Respiratory Abnormalities in Indoor Environments

This study investigates the use of various types of radar to recognize respiratory abnormalities in an indoor environment. Unlike prior approaches limited to fixed scenarios, our approach accommodates not only diverse measured angles and distances of the subject but also radar setups, exposing a Machine Learning method to a wide range of signal properties. Utilizing classical feature engineering and optimized hyperparameters, data from different radar types-Frequency Modulated Continuous Wave (FMCW) and Ultra-Wideband (UWB)—accurately classify respiratory behavior as normal or abnormal. This achieves a high accuracy of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 8. 1 7 \%}$</tex> on a test dataset comprising diverse respiratory patterns from various radar setups and target positions.