Synthetic aperture radar （SAR） systems have become an important tool for fine-resolution mapping and other remote sensing operations. The multi-channel SAR ground moving-target indication （GMTI） must process its d...Synthetic aperture radar （SAR） systems have become an important tool for fine-resolution mapping and other remote sensing operations. The multi-channel SAR ground moving-target indication （GMTI） must process its data to produce not only the image of surveillance area but also the information of the ground moving-targets. The topic of moving-target detection in clutter has been extensively studied, and there are many methods that are used to detect moving targets, such as displaced phase center antenna （DPCA） method, along-track interfero-metric （ATI） phase, space-time adaptive processing （STAP）, or some other metrics. A canonical framework is proposed that encompasses all the multi-channel SAR-GMT methods, namely, DPCA and ATI. The statistical test metric for multi-channel SAR-GMTI is established in a simple form, via the definition of the complex central Wishart distribution, to deduce the statistics of the test metric, and the probability distribution of the test metric for multichannel SAR-GMTI has the complex central Wishart distribution of 1×1 case, namely the X^2 distribution. The theory foundation offers the possibility to construct the united multi-channel SAR-GMTI detector, and derives the constant false alarm rate （CFAR） detector tests for separating moving targets from clutter.展开更多
稳态视觉诱发电位(Steady State Visual Evoked Potential,SSVEP)凭借其信噪比高、信息传输率高等优点成为脑控技术主流范式之一。对SSVEP信号的特征识别和特征提取算法是SSVEP系统研究的关键问题,但目前研究中适用于SSVEP算法的综述较...稳态视觉诱发电位(Steady State Visual Evoked Potential,SSVEP)凭借其信噪比高、信息传输率高等优点成为脑控技术主流范式之一。对SSVEP信号的特征识别和特征提取算法是SSVEP系统研究的关键问题,但目前研究中适用于SSVEP算法的综述较少。针对此问题,总结近年来适用于SSVEP机器学习算法,从机器学习的角度将算法分为无监督学习和有监督学习,介绍典型相关分析、卷积神经网络等算法的原理和适用范围。总结当前SSVEP算法在实际应用中的不足之处,并讨论SSVEP所面临的机遇与挑战。展开更多
基金the National Natural Science Foundation of China (60472097 and 60502054)
文摘Synthetic aperture radar （SAR） systems have become an important tool for fine-resolution mapping and other remote sensing operations. The multi-channel SAR ground moving-target indication （GMTI） must process its data to produce not only the image of surveillance area but also the information of the ground moving-targets. The topic of moving-target detection in clutter has been extensively studied, and there are many methods that are used to detect moving targets, such as displaced phase center antenna （DPCA） method, along-track interfero-metric （ATI） phase, space-time adaptive processing （STAP）, or some other metrics. A canonical framework is proposed that encompasses all the multi-channel SAR-GMT methods, namely, DPCA and ATI. The statistical test metric for multi-channel SAR-GMTI is established in a simple form, via the definition of the complex central Wishart distribution, to deduce the statistics of the test metric, and the probability distribution of the test metric for multichannel SAR-GMTI has the complex central Wishart distribution of 1×1 case, namely the X^2 distribution. The theory foundation offers the possibility to construct the united multi-channel SAR-GMTI detector, and derives the constant false alarm rate （CFAR） detector tests for separating moving targets from clutter.
文摘稳态视觉诱发电位(Steady State Visual Evoked Potential,SSVEP)凭借其信噪比高、信息传输率高等优点成为脑控技术主流范式之一。对SSVEP信号的特征识别和特征提取算法是SSVEP系统研究的关键问题,但目前研究中适用于SSVEP算法的综述较少。针对此问题,总结近年来适用于SSVEP机器学习算法,从机器学习的角度将算法分为无监督学习和有监督学习,介绍典型相关分析、卷积神经网络等算法的原理和适用范围。总结当前SSVEP算法在实际应用中的不足之处,并讨论SSVEP所面临的机遇与挑战。
