We introduce and develop a novel approach to outlier detection based on adaptation of random subspace learning. Our proposed method handles both high-dimension low-sample size and traditional low-dimensional high-samp...We introduce and develop a novel approach to outlier detection based on adaptation of random subspace learning. Our proposed method handles both high-dimension low-sample size and traditional low-dimensional high-sample size datasets. Essentially, we avoid the computational bottleneck of techniques like Minimum Covariance Determinant (MCD) by computing the needed determinants and associated measures in much lower dimensional subspaces. Both theoretical and computational development of our approach reveal that it is computationally more efficient than the regularized methods in high-dimensional low-sample size, and often competes favorably with existing methods as far as the percentage of correct outlier detection are concerned.展开更多
在高动态与强干扰条件下,卫星导航接收机码跟踪环路与载波跟踪环路会出现异常抖动进而影响卫星导航接收机的授时精度,针对这一问题提出了一种新的基于抗野值Kalman滤波技术的卫星导航接收机授时方法。该方法可以有效消除码环与载波环的...在高动态与强干扰条件下,卫星导航接收机码跟踪环路与载波跟踪环路会出现异常抖动进而影响卫星导航接收机的授时精度,针对这一问题提出了一种新的基于抗野值Kalman滤波技术的卫星导航接收机授时方法。该方法可以有效消除码环与载波环的异常抖动对解算出的卫星钟差的影响,并且对连续出现的野值也有很好的剔除效果。同时该授时方法还可以有效地对卫星导航接收机的晶振频率误差进行估计,进而对1PPS(pulse per second)信号发生器的频率控制字进行调整,提高系统的授时精度。经过实验验证,该授时方法可以有效提高卫星导航接收机的授时精度以及授时系统的鲁棒性。展开更多
随着高频开关技术的发展,大量电力电子设备的应用除导致了低频谐波发射外,也引发了2~150 kHz范围内超高次谐波发射现象,由此所带来的高频电磁干扰等问题越发突出。超高次谐波信号分布频带宽、幅值小,实际工程应用中对采样频率要求较高,...随着高频开关技术的发展,大量电力电子设备的应用除导致了低频谐波发射外,也引发了2~150 kHz范围内超高次谐波发射现象,由此所带来的高频电磁干扰等问题越发突出。超高次谐波信号分布频带宽、幅值小,实际工程应用中对采样频率要求较高,产生的数据量大,现有检测算法难以同时兼顾低数据存储量与高频域分辨率的需求。针对这一矛盾,利用超高次谐波信号的离群点特性,在偏态分布模型的基础上,结合参数自适应的基于密度的噪声应用空间聚类(Density-based Spatial Clustering of Applications with Noise,DBSCAN)算法,提出了一种新的超高次谐波精确量化算法(Skewed Distribution Density-based Spatial Clustering of Applications with Noise,SD-DBSCAN)。该算法检测结果分辨率可达5 Hz,且数据量低于原始频谱的0.05%,可同时满足高分辨率和低数据量的要求。最后,通过仿真和实测平台对所提算法的有效性进行了论证。基于所提算法,可探究并形成监控及测量超高次谐波发射影响的全新方法和技术,为超高次谐波领域的研究提供理论依据和实践经验。展开更多
文摘We introduce and develop a novel approach to outlier detection based on adaptation of random subspace learning. Our proposed method handles both high-dimension low-sample size and traditional low-dimensional high-sample size datasets. Essentially, we avoid the computational bottleneck of techniques like Minimum Covariance Determinant (MCD) by computing the needed determinants and associated measures in much lower dimensional subspaces. Both theoretical and computational development of our approach reveal that it is computationally more efficient than the regularized methods in high-dimensional low-sample size, and often competes favorably with existing methods as far as the percentage of correct outlier detection are concerned.
文摘在高动态与强干扰条件下,卫星导航接收机码跟踪环路与载波跟踪环路会出现异常抖动进而影响卫星导航接收机的授时精度,针对这一问题提出了一种新的基于抗野值Kalman滤波技术的卫星导航接收机授时方法。该方法可以有效消除码环与载波环的异常抖动对解算出的卫星钟差的影响,并且对连续出现的野值也有很好的剔除效果。同时该授时方法还可以有效地对卫星导航接收机的晶振频率误差进行估计,进而对1PPS(pulse per second)信号发生器的频率控制字进行调整,提高系统的授时精度。经过实验验证,该授时方法可以有效提高卫星导航接收机的授时精度以及授时系统的鲁棒性。
文摘随着高频开关技术的发展,大量电力电子设备的应用除导致了低频谐波发射外,也引发了2~150 kHz范围内超高次谐波发射现象,由此所带来的高频电磁干扰等问题越发突出。超高次谐波信号分布频带宽、幅值小,实际工程应用中对采样频率要求较高,产生的数据量大,现有检测算法难以同时兼顾低数据存储量与高频域分辨率的需求。针对这一矛盾,利用超高次谐波信号的离群点特性,在偏态分布模型的基础上,结合参数自适应的基于密度的噪声应用空间聚类(Density-based Spatial Clustering of Applications with Noise,DBSCAN)算法,提出了一种新的超高次谐波精确量化算法(Skewed Distribution Density-based Spatial Clustering of Applications with Noise,SD-DBSCAN)。该算法检测结果分辨率可达5 Hz,且数据量低于原始频谱的0.05%,可同时满足高分辨率和低数据量的要求。最后,通过仿真和实测平台对所提算法的有效性进行了论证。基于所提算法,可探究并形成监控及测量超高次谐波发射影响的全新方法和技术,为超高次谐波领域的研究提供理论依据和实践经验。
