本研究提出一種基於多普勒擴展目標（Doppler-Spread Target, DST）並結合先追蹤後偵測（Track-Before-Detect, TBD）之行人偵測演算法，以因應真實城市環境中雜波強烈且空間分布不均所造成之偵測困難。於此類場景中，接收訊號常同時包含地面、牆面、路側設施與各類固定反射體所造成之靜態回波，使行人目標回波功率偏弱易被雜波掩蔽而產生漏偵設情況。提出的演算法能夠抑制量測時雜訊所造成的干擾並提升行人目標的訊雜比並且在有虛警產生的情況下還是能夠很好的分辨出行人目標，從而提升行人偵測效能。本研究針對頻率調變連續波雷達進行了詳細的理論分析，包括訊號的數學模型以及在行人兩種不同移動情境進行探討，以訊號模擬與量測實驗，驗證DST-TBD演算法之有效性以及對行人目標偵測有顯著提升，並與其他演算法進行比較。本篇實驗使用 Two-Pulse MTI canceller 抑制零多普勒靜態雜波，降低背景能量抬升對門檻判決之影響；並採用文獻之 swarm orthogonal matching pursuit（sOMP）進行距離向稀疏重建，以提升距離可解析度與可檢出性。針對行人回波之多普勒展延特性，進一步以 DST 模型進行能量聚合，形成檢定統計量，並在固定虛警率 P_{fa}約束下結合 DST- SOCS CFAR產生候選點跡。為克服步態週期造成之單幀能量起伏與候選點最大值易受雜訊尖峰干擾之問題，將候選點跡輸入 TBD 架構，於固定跨幀視窗內進行分數累積與軌跡關聯，利用運動一致性約束抑制孤立假警與誤定位，輸出連續且合理之目標軌跡。最後與其他演算法進行比較證明所提出演算法的有效性。

This thesis proposes a pedestrian detection algorithm that combines a Doppler-Spread Target (DST) model with Track-Before-Detect (TBD) to improve detection in real urban environments with strong, spatially non-uniform clutter. Static echoes from the ground, walls, and roadside infrastructure often mask weak pedestrian returns and cause missed detections. The proposed method enhances pedestrian SNR and remains robust even under false alarms.
A detailed FMCW radar signal model is presented, and two pedestrian motion scenarios are investigated through both simulations and real measurements. In the processing chain, a two-pulse MTI canceller is used to suppress zero-Doppler clutter, and swarm OMP (sOMP) is adopted for sparse range reconstruction to improve range resolution. A DST model then aggregates Doppler-spread energy to form a detection statistic, and DST-SOCS CFAR is applied under a specified P_{fa}to generate candidate plots. To mitigate single-frame fluctuations and noise spikes, these candidates are further processed by TBD, which accumulates multi-frame scores and enforces motion-consistency constraints to suppress isolated false alarms and produce continuous, physically plausible tracks. Experimental comparisons with other algorithms demonstrate significant performance gains of the proposed DST–TBD approach.

[1] A. Pandharipande et al., "Sensing and Machine Learning for Automotive Perception: A Review," in IEEE Sensors Journal, vol. 23, no. 11, pp. 11097-11115, 1 June1, 2023, doi: 10.1109/JSEN.2023.3262134.
[2] Global Status Report on Road Safety 2018: Summary, World Health Organization, Geneva, Switzerland, 2018.
[3] F. Camara et al., "Pedestrian Models for Autonomous Driving Part I: Low-Level Models, From Sensing to Tracking," in IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 10, pp. 6131-6151, Oct. 2021, doi: 10.1109/TITS.2020.3006768.
[4] L. Chen et al., "Deep Neural Network Based Vehicle and Pedestrian Detection for Autonomous Driving: A Survey," in IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 6, pp. 3234-3246, June 2021, doi: 10.1109/TITS.2020.2993926.
[5] A. Lazaro, M. Lazaro, R. Villarino and P. De Paco, "New Radar Micro-Doppler Tag for Road Safety Based on the Signature of Rotating Backscatters," in IEEE Sensors Journal, vol. 21, no. 6, pp. 8604-8612, 15 March15, 2021, doi: 10.1109/JSEN.2020.3048081.
[6] Z. Zhang, X. Wang, D. Huang, X. Fang, M. Zhou and Y. Zhang, "MRPT: Millimeter-Wave Radar-Based Pedestrian Trajectory Tracking for Autonomous Urban Driving," in IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1-17, 2022, Art no. 8000117, doi: 10.1109/TIM.2021.3139658.
[7] E. Marchetti, “Comparison of pedestrian reflectivities at 24 and 300 GHz,” in Proc. 18th Int. Radar Symp. (IRS), Jun. 2017, pp. 1–7.
[8] D. Belgiovane, C.-C. Chen, M. Chen, S. Y.-P. Chien, and R. Sherony, “77 GHz radar scattering properties of pedestrians,” in Proc. IEEE Radar Conf., May 2014, pp. 0735–0738.
[9] S. Zang, M. Ding, D. Smith, P. Tyler, T. Rakotoarivelo, and M. A. Kaafar, “The impact of adverse weather conditions on autonomous vehicles: How rain, snow, fog, and hail affect the performance of a self-driving car,” IEEE Veh. Technol. Mag., vol. 14, no. 2, pp. 103–111, Jun. 2019.
[10] C. Xu, F. Wang, Y. Zhang, L. Xu, M. Ai, and G. Yan, “Two-level CFAR algorithm for target detection in mmWave radar,” in Proc. Int. Conf. Comput. Eng. Appl. (ICCEA), Jun. 2021, pp. 240–243.
[11] H. M. Finn, “Adaptive detection mode with threshold control as a function of spatially sampled clutter level estimates,” Rome Air Develop. Center (RADC), Rome, NY, USA, Tech. Rep. RADC-TR-73-61, 1968.
[12] M. Weiss, “Analysis of some modified cell-averaging CFAR processors in multiple-target situations,” IEEE Trans. Aerosp. Electron. Syst., vols. AES-18, no. 1, pp. 102–114, Jan. 1982.
[13] V. Hansen and J. Sawyers, “Detectability loss due to ‘Greatest of’ selection in a cell-averaging CFAR,” IEEE Trans. Aerosp. Electron. Syst., vols. AES-16, no. 1, pp. 115–118, Jan. 1980.
[14] X. Fang, J. Li, Z. Zhang and G. Xiao, "FMCW-MIMO Radar-Based Pedestrian Trajectory Tracking Under Low- Observable Environments," in IEEE Sensors Journal, vol. 22, no. 20, pp. 19675-19687, 15 Oct.15, 2022, doi: 10.1109/JSEN.2022.3203154.
[15] D. Belgiovane and C. -C. Chen, "Micro-Doppler characteristics of pedestrians and bicycles for automotive radar sensors at 77 GHz," 2017 11th European Conference on Antennas and Propagation (EUCAP), Paris, France, 2017, pp. 2912-2916, doi: 10.23919/EuCAP.2017.7928457.
[16] V. C. Chen, F. Li, S. . -S. Ho and H. Wechsler, "Micro-Doppler effect in radar: phenomenon, model, and simulation study," in IEEE Transactions on Aerospace and Electronic Systems, vol. 42, no. 1, pp. 2-21, Jan. 2006, doi: 10.1109/TAES.2006.1603402.
[17] W. Zhang, H. Li, G. Sun and Z. He, "Enhanced Detection of Doppler-Spread Targets for FMCW Radar," in IEEE Transactions on Aerospace and Electronic Systems, vol. 55, no. 4, pp. 2066-2078, Aug. 2019, doi: 10.1109/TAES.2019.2925433.
[18] Z. Zhao, H. Wang, L. Cao, D. Wang and C. Fu, "Doppler-Spread Target Summation Variability Index CFAR Detector for FMCW Radar," in IEEE Sensors Journal, vol. 24, no. 20, pp. 32519-32532, 15 Oct.15, 2024, doi: 10.1109/JSEN.2024.3432179.
[19] R. Zhang, H. Yang, F. Höflinger and L. M. Reindl, "Adaptive Zero Velocity Update Based on Velocity Classification for Pedestrian Tracking," in IEEE Sensors Journal, vol. 17, no. 7, pp. 2137-2145, 1 April1, 2017, doi: 10.1109/JSEN.2017.2665678.
[20] L. Wang, Z. Ai, Q. Wang, J. Wang and X. Zhang, "Swarm Sparse Range Profile Recovery From Fewer Measurements of FMCW Radar," in IEEE Transactions on Aerospace and Electronic Systems, vol. 60, no. 5, pp. 7056-7074, Oct. 2024, doi: 10.1109/TAES.2024.3411566
[21] H. M. Finn, “Adaptive detection mode with threshold control as a function of spatially sampled clutter level estimates,” Rome Air Develop. Center (RADC), Rome, NY, USA, Tech. Rep. RADC-TR-73-61, 1968.
[22] M. Weiss, “Analysis of some modified cell-averaging CFAR processors in multiple-target situations,” IEEE Trans. Aerosp. Electron. Syst., vols. AES-18, no. 1, pp. 102–114, Jan. 1982.
[23] V. Hansen and J. Sawyers, “Detectability loss due to ‘Greatest of’ selection in a cell-averaging CFAR,” IEEE Trans. Aerosp. Electron. Syst., vols. AES-16, no. 1, pp. 115–118, Jan. 1980.
[24] H. Rohling, “Radar CFAR thresholding in clutter and multiple target situations,” IEEE Trans. Aerosp. Electron. Syst., vols. AES-19, no. 4, pp. 608–621, Jul. 1983.
[25] E. Grossi, M. Lops and L. Venturino, "A Novel Dynamic Programming Algorithm for Track-Before-Detect in Radar Systems," in IEEE Transactions on Signal Processing, vol. 61, no. 10, pp. 2608-2619, May15, 2013, doi: 10.1109/TSP.2013.2251338.
[26] S. Abdulatif, F. Aziz, B. Kleiner and U. Schneider, "Real-time capable micro-Doppler signature decomposition of walking human limbs," 2017 IEEE Radar Conference (RadarConf), Seattle, WA, USA, 2017, pp. 1093-1098, doi: 10.1109/RADAR.2017.7944367.
[27] P. Held et al., "Radar-Based Analysis of Pedestrian Micro-Doppler Signatures Using Motion Capture Sensors," 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 2018, pp. 787-793, doi: 10.1109/IVS.2018.8500656.
[28] R. Veld, A. B. J. Kokkeler, A. Chiumento and Y. Miao, "Height-Adaptive Human Gait Model Through Radar-Driven Pipeline With Two Co-Located mmWave MIMO Radars," in IEEE Access, vol. 12, pp. 7199-7223, 2024, doi: 10.1109/ACCESS.2024.3350354.
[29] IWR1443BOOST Evaluation Module mmWave Sensing Solution: User's Guide. Dallas, TX, USA : Texas Instruments, May 2017, revised May 2020. [Online] Available: https://www.ti.com.cn/cn/lit/ug/swru518d/swru518d.pdf