System-Level Design and Implementation of an Integrated FMCW LiDAR : Optimization of Sweep Linearity and Detection Performance
통합형 FMCW LiDAR의 시스템 레벨 설계 및 구현
- 주제(키워드) FMCW , LiDAR , Laser Diode , FPGA , Hilbert Transform , FFT , Kernel , PLL , TIA , Optics
- 발행기관 서강대학교 일반대학원
- 지도교수 Jinwook Burm
- 발행년도 2026
- 학위수여년월 2026. 2
- 학위명 석사
- 학과 및 전공 일반대학원 전자공학과
- 실제URI http://www.dcollection.net/handler/sogang/000000082563
- UCI I804:11029-000000082563
- 본문언어 영어
- 저작권 논문은 저작권에 의해 보호받습니다.
초록(요약문)
Frequency-Modulated Continuous-Wave (FMCW) LiDAR has gained attention as a next-generation sensing solution due to its capability for coherent detection, immunity to ambient noise, and simultaneous distance– velocity measurement. However, practical deployment is hindered by challenges in cost, real-time performance, and analog hardware complexity. This work presents three FMCW LiDAR models, each targeting a different design constraint. The first model is optimized for low-cost implementation, minimizing analog and digital complexity while preserving essential functionality. The second model is designed for real-time signal processing, incorporating FFT and MAC operations within an FPGA-based pipeline for high-throughput distance computation. The third model utilizes EO-PLL and a pulse-counting architecture to eliminate ADC–DSP chains, improving power efficiency and robustness. Through comparative analysis, this study examines trade-offs across complexity, accuracy, and performance. The proposed architectures provide scalable FMCW LiDAR solutions, with each model suited for specific application needs ranging from cost-sensitive platforms to high-performance embedded systems.
more목차
Abstract 1
Chapter 1. Introduction 2
Chapter 2. FMCW LiDAR Overview 5
2.1 Frequency Modulation (Linearization) 5
2.2 Continuous Wave (Demodulation) 7
Chapter 3. Design Ver 1 (for Low Cost System) 9
3.1 Degraded TX System 9
3.2 Correction Method 11
3.2.1 TX Pre-Distortion (Based on Resampling Logic) 11
3.2.2 RX Resampling 13
3.3 Effectiveness of Combining two Method 15
3.4 Final Ranging Performance (ver1) 18
Chapter 4. Design Ver 2 (for Real-Time ADAS) 27
4.1 Base TX System 27
4.2 Custom AFE 29
4.3 Nonlinearity Tracking 32
4.4 Final Ranging Performance (ver2) 35
Chapter 5. Techniques for Circuit-Level Design 41
5.1 Electro-Optical Phase Locked Loop 41
5.2 Pulse_Counting Method 45
Chapter 6. Conclusion 48
References 50
List of Figures & Tables
Figure 1.1 ToF / FMCW LiDAR Comparison 4
Figure 2.1 FMCW LiDAR Total ranging system 6
Figure 2.2 Frequency Chirping 7
Figure 3.1 FMCW LiDAR Total ranging system (Double Correction) 12
Figure 3.2 FFT spectra of distance signals at 1 m, 2 m, and 3 m 14
Figure 3.3 Distance variation under temperature drift 17
Figure 3.4 Effect of double calibration on distance spectra 19
Figure 3.5 FWHM of the beat spectrum at a target distance of 1 m 20
Figure 3.6 Error rate and STD against Distance after correction 21
Figure 3.7 Histograms of measurement deviations at 0.5m & 1.0m 22
Figure 3.8 Measured displacement for fine motion around 0.5m 23
Figure 4.1 Overview of the ECLD 28
Figure 4.2 Overall Structure of the proposed AFE 30
Figure 4.3 The active feedback circuit for canceling input dc 31
Figure 4.4 Post-Amplifier with CMFB for gain boosting and output buffer
for measurement 31
Figure 4.5 HDL architecture of the proposed kernel tracking 33
Figure 4.6 STD Comparison: AUX-Based vs Stored-Kernel 34
Figure 4.7 Experimental setup of the proposed FMCW LiDAR system · 36
Figure 4.8 Experimental results of total ranging 38
Figure 4.9 Distribution of measurement errors 39
Figure 5.1 EO-PLL TX System Block Diagram 42
Figure 5.2 EO-PLL TX System Circuit 43
Figure 5.3 Proposed Pulse Counting Architecture and AFE Circuit 45
Figure 5.4 Overall Architecture of the Proposed FMCW LiDAR Transceiver 47
Table 1. Initial Nonlinearity 10
Table 2. Performance Summary and Comparison (ver1) 26
Table 3. Performance Summary and Comparison (ver2) 40
