复杂地面杂波环境中LFMCW体制雷达慢速弱目标探测技术研究

With small radar cross section and low Doppler velocity overlapped with ground clutter, slow moving weak ground target detection has continued to be a challenge for modern radar applications. In this project, linear frequency modulated continuous wave (LFMCW) radar for slow moving ground targets detection is investigated. The distribution and time-varying parameters of ground clutter in range bins are estimated. The clutter estimates are then used to design LFMCW waveform adaptively for transmission to minimize the out-of-bin clutter contributions to the range-Doppler bins of interest, such that the signal-to-clutter ratio (SCR) of target echo in the bins of interest is improved. Human movements such as walking, strolling or creeping, are modeled. According to the movement models, the Doppler frequency of walking human echo is time-varying, a time-frequency analysis algorithm is used to obtain the micro-Doppler characteristics of human movements. According to the Doppler characteristics, the echo of LFMCW radar is analyzed to form a range-Doppler region, employing a fitting or pattern recognition algorithm, the micro-Doppler curve of potential human target is extracted. Combining a clutter distribution recognition algorithm and interfering-targets censoring method, a constant false alarm rate detector is designed for complex clutter environments and multiple targets situations. The range-Doppler cells in the extracted Doppler curve are used by the CFAR detector to determine whether a target exists or not. A LFMCW radar system for slow moving ground target has the advantages of small size, light weight, high range resolution, no blind range and low price, and is especially suitable for military and civilian areas, such as battlefield surveillance, anti-terrorism and disaster rescue.

由于雷达散射截面较小，且多普勒频率与地物杂波重叠，强地物杂波环境中慢速弱目标探测一直是雷达的难题。本课题采用线性调频连续波体制雷达对慢速弱目标探测开展研究。通过地物杂波分布类型与参数估计，设计自适应线性调频连续波发射波形，减小杂波在感兴趣距离-多普勒区域能量分布，提高目标回波信杂比；建立人体运动物理模型，探索人体运动微多普勒变化规律，构建雷达回波时间-多普勒二维图，采用匹配或模式识别算法提取目标多普勒曲线，获得慢速人体目标可能点迹；在多普勒曲线提取基础上，将杂波分布识别与干扰目标剔除相结合，设计针对低多普勒区域杂波变化剧烈和多目标场景的恒虚警检测策略，降低雷达探测虚警概率，提高对弱目标检测概率。具备慢速弱目标探测能力的LFMCW体制雷达具有体积小、重量轻、距离分辨力高、无距离盲区和成本低的优点，特别适用于战场前沿侦察、重点区域监视与灾区人员搜救等特殊环境，在军用和民用领域将发挥重要作用。

线性调频连续波（LFMCW）体制雷达体积小、重量轻，结构简单，常用于近距离目标探测。本研究从雷达发射波形设计、信号处理与恒虚警检测三个角度出发，来研究提高LFMCW体制地面监视雷达探测慢速弱目标能力的方法。.在波形设计方面，设计了基于时域参差编码和双曲线编码的SFLFM脉冲串正交波形集。通过对波形中子脉冲发射时刻进行时域参差编码，选用子脉冲时宽、带宽和步进频率经过合理设计的SFLFM脉冲串作为发射波形，并对各波形中子脉冲发射顺序进行双曲线编码，使得波形集具有较低的自相关旁瓣峰值和互相关峰值。.在人体运动回波建模与信号处理方面，建立LFMCW雷达人体运动测试场景，简化较为复杂的Boulic人体运动模型，获得行进人体LFMCW雷达总回波，设计了基于人体运动模型的相位补偿积累方法，提高了多普勒扩散人体回波的能量积累效率，从而提升了检测信噪比。.在恒虚警检测方面，设计了针对多目标场景的恒虚警检测器SKMR-CFAR，该检测器利用韦布尔杂波对数变换后偏斜度为定值，并结合统计量均值比来判断环境中是否存在干扰目标和杂波边缘。该方法有效缓解了干扰目标和杂波边缘对检测性能的影响，具有很强的干扰目标和杂波边缘判断能力，从而提高了LFMCW雷达在非均匀杂波环境中对弱目标的检测性能。