Accurate Large-Scale Motion Sensing With FMCW Radar Based on Range-Dependent DFT Technique

Frequency-modulated continuous wave (FMCW) radars are widely employed for motion sensing. However, when the target undergoes large-scale motion, it often spans multiple range bins, which introduces phase errors into the demodulation process and prevents the realization of high-precision motion recovery. To address this challenge, a range-dependent Discrete Fourier Transform (DFT) technique is proposed, where the raw intermediate frequency (IF) signals are processed using DFT with varying points, determined based on the target’s range. Through this technique, the range bins traversed by motion are aligned into a single bin, effectively eliminating phase errors caused by range bin transitions. Simulations and experiments were conducted to evaluate the effectiveness of the proposed technique, and the results demonstrate that the root mean square error (RMSE) of motion recovery was reduced by a factor of 90, underscoring its significance for large-scale motion sensing.