Fusion Dynamic Time Warping Algorithm for Hand Gesture Recognition in FMCW Radar System

In this paper, we propose a fusion dynamic time warping (FDTW) method for hand gesture recognition using a frequency modulated continuous wave (FMCW) radar. First, we use two-dimensional Fast Fourier Transform (2D-FFT) to estimate the parameters of the radar raw data, and construct the range-time spectrogram (RTM) and Doppler-time spectrogram (DTM). Second, a central time–frequency trajectory is clustered for each hand gesture spectrogram using the k-means algorithm, and then the hand gesture is recognized by the proposed Fusion Dynamic Time Warping (FDTW) algorithm. Experiments with radar data show that the proposed algorithm can significantly improve the recognition accuracy with the time complexity reduced by more than 50%.