Discriminative Fusion Methods for FMCW LiDAR: Leveraging All-phase FFT and Time-shifted Phase Difference

Challenges arise in the frequency modulated continuous wave (FMCW) Laser detection and ranging (LiDAR) backend signal processing FFT frequency estimation algorithm, including the problems of picket fence effect, spectrum leakage, and processing speed limitation. These challenges hinder the accuracy and speed of target detection. This paper proposes a high-precision and fast processing method that combines digital down-conversion (DDC) with all-phase FFT (apFFT) and time-shifted phase difference method. The method effectively combines the phase characteristics of DDC output signals with the phase invariance property of the apFFT algorithm. Experimental results show the average ranging error accuracy of 0.55cm with 25 measurements at 94m, which is a 51.75% improvement over traditional FFT algorithms. Additionally, the time complexity is reduced by 54.17% at 4096 sampling points.