Double Resampling Architecture for Distance Measurement in FMCW LiDAR with Degraded Transmitter Systems

초록

Frequency Modulated Continuous Wave (FMCW) LiDAR systems require the precise generation of linear chirp signals for high-resolution distance measurements. While previous studies have primarily focused on the intrinsic frequency nonlinearity of laser diode, this work considers the broader impact of the entire transmit (TX) chain on ranging accuracy. The nonlinearity of the TX system is newly classified into a static component and a residual component, each requiring independent optimization strategies. To address this, we introduce a double-resampling approach that combines pre-distortion with a post-processing algorithm. Both stages employ similar Hilbert transform-based structures, minimizing algorithmic complexity while enabling high-precision operation even with low-cost or aged TX hardware, which demonstrates strong commercial potential. Experimental results under free-space conditions show a standard deviation (STD) of 0.18 mm and an error rate (Accuracy) of 0.17% at 1 m, with a full width at half maximum (FWHM) of 6.28 mm, indicating the distance resolution.

키워드