Timing jitter is one of the main factors that influence on the accuracy of time domain precision measurement. Timing jitter compensation is one of the problems people concern. Because of the flaws of median method, PD...Timing jitter is one of the main factors that influence on the accuracy of time domain precision measurement. Timing jitter compensation is one of the problems people concern. Because of the flaws of median method, PDF deconvolution method and synthetic method, we put forward a new method for timing jitter compensation, which is called advanced median method. The theory of the advanced median method based on probability and statistics is analyzed, and the process of the advanced median method is summarized in this paper. Simulation and experiment show that compared with other methods, the new method could compensate timing jitter effectively.展开更多
基金National Natural Science Foundation of China (50605065)
文摘Timing jitter is one of the main factors that influence on the accuracy of time domain precision measurement. Timing jitter compensation is one of the problems people concern. Because of the flaws of median method, PDF deconvolution method and synthetic method, we put forward a new method for timing jitter compensation, which is called advanced median method. The theory of the advanced median method based on probability and statistics is analyzed, and the process of the advanced median method is summarized in this paper. Simulation and experiment show that compared with other methods, the new method could compensate timing jitter effectively.
文摘基于目标阴影的跟踪技术是视频合成孔径雷达(video synthetic aperture radar,ViSAR)目标探测的重要手段,但ViSAR数据存在目标特征不明显且随时间不规则变化、相干斑噪声干扰强等问题,使得ViSAR目标阴影跟踪精度较低。为此,提出了一种鲁棒的基于时间信息加权的ViSAR目标阴影跟踪算法。针对目标特征不明显且随时间不规则变化的问题,将尺度自适应均值偏移(adaptive scale mean shift,ASMS)跟踪算法引入到ViSAR目标阴影跟踪中,同时在ASMS算法的背景比例加权(background ratio weighted,BRW)技术中添加历史帧的特征,并对尺度正则项进行时间信息加权修正,来对目标特征进行整合。针对相干斑噪声干扰强的问题,对ASMS算法加入局部中值滤波操作的预处理步骤,在不增加计算量的同时平滑了噪声。在对两类ViSAR数据集上不同尺寸、不同运动状态的目标阴影的跟踪实验结果表明,与现有高性能跟踪算法相比,所提算法在保证了实时性的基础上提高了跟踪精度,且不需要额外的训练样本,具备较好的工程应用价值。