Hand Gesture Recognition via MIMO Radar Sensors and Space-Frequency Domain Information

With the development of radar systems, hand gesture recognition of radar sensors has become easier to operate, and the resolution has increased. Nowadays, radar gesture recognition frequency employs (multiple-input and multiple-output) MIMO radar as a sensor since it has a better spatial resolution. This article proposes a hand gesture recognition based on a MIMO radar sensor, which can differentiate five gestures: swipe left, swipe right, pat, push, and pull. After receiving the data from the radar sensor, we first apply a two-dimensional Fast-Fourier Transform (2D-FFT) for a time series of range-Doppler maps. Next, we detect the moving target range, velocity, and angle values through time. Finally, the classification and regression tree (CART) algorithm is applied to a collected dataset, with targets’ time-variant characteristics as the parameters. The overall recognition rate of 94% is obtained from the proposed system using a decision-tree-based classifier. The experiment results show that this gesture-recognition system is promising in classifying multiple gestures.