Signal Processing Chain with Spiking Neural Networks for Radar-based Gesture Sensing

Gesture sensing (GS) technologies enable the creation of touchless, non-invasive, and intuitive interfaces that allow users to interact with various devices without thThe impact of post-training quantization on accuracyneed for physical contact or manual controls [51]. It has enabled a wide range of applications, including automotive, gaming, TVs, and smartphones. Traditionally, GS technology is dominated by vision-based systems where computer vision techniques are applied to rich datasets obtained with camera sensors [52–55]. Camera-based solutions, however, raise privacy concerns since they operate with photos and videos. Additionally, they require appropriate lighting conditions and weather conditions (in outdoor scenarios), which limits their application.