检索结果-南通市图书馆

Simplified unscented particle filter for nonlinear/non-Gaussian Bayesian estimation

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Engineering and Electronics 2013年第3期24卷 537-544页

作者： Junyi Zuo Yingna Jia quanxue gao School of Aeronautics Northwestern Polytechnical University Aviation Equipment Research Institute Qing'an Group Corporation Limited State Key Laboratory of Integrated Service Networks Xidian University

Particle filters have been widely used in nonlinear/non- Gaussian Bayesian state estimation problems. However, efficient distribution of the limited number of particles （n state space remains a critical issue in designing a particle filter. A simplified unscented particle filter （SUPF） is presented, where particles are drawn partly from the transition prior density （TPD） and partly from the Gaussian approximate posterior density （GAPD） obtained by a unscented Kalman filter. The ratio of the number of particles drawn from TPD to the number of particles drawn from GAPD is adaptively determined by the maximum likelihood ratio （MLR）. The MLR is defined to measure how well the particles, drawn from the TPD, match the likelihood model. It is shown that the particle set generated by this sampling strategy is more close to the significant region in state space and tends to yield more accurate results. Simulation results demonstrate that the versatility and es- timation accuracy of SUPF exceed that of standard particle filter, extended Kalman particle filter and unscented particle filter.

关键词： nonlinear filtering particle filter unscented Kalman filter importance density function.

Discriminant Neighborhood Structure Embedding Using Trace Ratio Criterion for Image Recognition

在线全文

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2015年第11期3卷 64-70页

作者： Jing Wang Fang Chen quanxue gao School of Telecommunications Engineering Xidian University Xi’an China State Key Laboratory of Integrated Services Networks Xidian University Xi’an China

Dimensionality reduction is very important in pattern recognition, machine learning, and image recognition. In this paper, we propose a novel linear dimensionality reduction technique using trace ratio criterion, namely Discriminant Neighbourhood Structure Embedding Using Trace Ratio Criterion (TR-DNSE). TR-DNSE preserves the local intrinsic geometric structure, characterizing properties of similarity and diversity within each class, and enforces the separability between different classes by maximizing the sum of the weighted distances between nearby points from different classes. Experiments on four image databases show the effectiveness of the proposed approach.

关键词： Dimensionality Reduction Manifold Learning Variability Trace Ratio

Laplacian Maximum Margin Criterion for Image Recognition

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2015年第11期3卷 58-63页

作者： Fang Chen Jing Wang quanxue gao State Key Laboratory of Integrated Services Networks Xidian University Xi’an China School of Telecommunications Engineering Xidian University Xi’an China

Previous works have demonstrated that Laplacian embedding can well preserve the local intrinsic structure. However, it ignores the diversity and may impair the local topology of data. In this paper, we build an objective function to learn the local intrinsic structure that characterizes both the local similarity and diversity of data, and then combine it with global structure to build a scatter difference criterion. Experimental results in face recognition show the effectiveness of our proposed approach.

关键词： Laplacian Embedding Local Intrinsic Structure Global Structure Face Recognition

Secrecy Transmission of NOMA-OTFS Based Multicast-Unicast Streaming

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2023年第1期20卷 1-13页

作者： Junfan Hu Jia Shi Xianyu Wang Xiaoju Lu Zan Li Zhuangzhuang Tie quanxue gao The State Key Laboratory of Integrated Services Networks Xidian UniversityXi’an 710071China Academy of Space Electronic Information Technology Xi’an 710100China Unit 69036 of PLA Urumqi 830000China

This paper investigates the security performance of a cooperative multicast-unicast system,where the users present the feature of high ***,we develop the non-orthogonal multiple access(NOMA)based orthogonal time frequency space(OTFS)transmission scheme,namely NOMAOTFS,in order to combat Doppler effect as well as to improve the spectral ***,we propose a power allocation method addressing the trade-off between the reliability of multicast streaming and the confidentiality of unicast *** on that,we utilize the relay selection strategy,to improve the security of unicast *** the context of multicastunicast streaming,our simulation findings validate the effectiveness of the NOMA-OTFS based cooperative transmission,which can significantly outperform the existing NOMA-OFDM in terms of both reliability and security.

关键词： OTFS NOMA physical layer security multicast-unicast

在线全文

学校读者我要写书评

暂无评论

结合图像信息的快速点云拼接算法

测绘学报 2016年第1期45卷 96-102页

作者：王瑞岩姜光高全学西安电子科技大学通信工程学院综合业务网理论及关键技术国家重点实验室陕西西安710071

现有三维激光扫描设备通常配有一个同轴相机,它可以对扫描场景进行拍摄。针对带有同轴相机的激光扫描设备,本文提出了一种结合图像信息的快速点云拼接算法。与传统拼接算法同时计算点云间的旋转和平移变换不同,本文对这两种变换分别进... 详细信息

现有三维激光扫描设备通常配有一个同轴相机,它可以对扫描场景进行拍摄。针对带有同轴相机的激光扫描设备,本文提出了一种结合图像信息的快速点云拼接算法。与传统拼接算法同时计算点云间的旋转和平移变换不同,本文对这两种变换分别进行求解。其中,不同扫描点云间的旋转变换是利用视觉几何知识由同轴相机在不同扫描站点下拍摄的图像直接获得,而平移变换是由本文提出的改进ICP算法得到。在改进的ICP算法中,只有平移变换的3个未知量被迭代计算,其输入是去除旋转变换后的点云。试验结果表明利用图像获得的点云旋转变换具有很高的准确性;并且由于本文算法中迭代过程只针对平移变换的3个变量进行计算,因此与需要迭代计算6个变量的传统ICP算法相比,本文算法计算复杂度大幅降低,同时更易收敛于全局最优值且收敛速度有所提高。

关键词：点云拼接图像信息视觉几何 ICP算法

Digital System Fault Diagnosis Based on BP Neural Network

在线全文

学校读者我要写书评

暂无评论

Digital System Fault Diagnosis Based on BP Neural Network

A New Center Position Control Implement

第二届仪器科学与技术国际学术研讨会

作者： Tian Qichuan, Pan Quan, gao Diguang, Han Rucheng, gao quanxue Taiyuan Heavy Machinery Inst. Northwestern Ploytechnical Univ.

＜正＞ In this paper, a fault diagnosis strategy of digital system based on BP neural network is presented. After training the neural network can be used to diagnose faults of digital system. Considering tradition BP algorithm of neural network may make weights run into local minimum, A modified algorithm are developed using BP algorithm and applying Metropolis criterion. Finally, this fault diagnosis system has been used in a digital system, the simulation setup is fabricated and tested, and the test results are compared with simulation results. Using this algorithm the approximation error of neural network is very small, and it is an effective approach and efficient in learning. In a conclusion , BP neural network is a good way to complex digital system fault diagnosis.

关键词： BP neural network fault diagnosis digital system fault code Metropolis criterion

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

A New Center Position Control Implement

第二届仪器科学与技术国际学术研讨会

作者： Tian Qichuan, Pan Quan, gao quanxue, Han Rucheng, gao Diguang Taiyuan Heavy Machinery Inst. Northwestern Ploytechnical Univ.

＜正＞ In many industry-producing fields, CPC （Center Position Control） system is needed to realize automatic control. In this paper, a new CPC implement is presented. It is designed based on single chip microcomputer solving the problem to minimize the quantity of edge-cut. In order to get high precision, new typical high accuracy sensors-CCD （Charge Coupled Device） has been adopted, and a self-adapting adjust hydraulic pressure clique has been used to provide appropriate control. In control algorithm, because the plant is difficult to modeling, a fuzzy control algorithm is used in the system and several parameters are designed to adjust according to the actual running effect in producing locale. Based on this designed, this control system can be used in others similar fields, It has many advantages such as robust, good reliability, high precision, and easy to adjust parameters. Comparison with other control is shown by experimental results and confirms the effectiveness of this control scheme. It has high value being generalized.

关键词： center position control fuzzy control charge coupled device hydraulic pressure system steel belt

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

结合自适应稀疏表示和全变分约束的图像重建

西安电子科技大学学报 2016年第1期43卷 12-17,109页

作者：王勇冯唐智陈楚楚乔倩倩杨笑宇王国栋高全学西安电子科技大学电子工程学院陕西西安710071 西安电子科技大学通信工程学院陕西西安710071

针对以二维小波变换和离散余弦变换为代表的固定正交基在图像压缩感知高分辨率重建中的局限性,提出了一种新的自适应冗余字典稀疏表示结合全变分约束的图像高分辨率重建算法.该算法以迭代过程的中间图像作为训练样本,通过自适应学习获... 详细信息

针对以二维小波变换和离散余弦变换为代表的固定正交基在图像压缩感知高分辨率重建中的局限性,提出了一种新的自适应冗余字典稀疏表示结合全变分约束的图像高分辨率重建算法.该算法以迭代过程的中间图像作为训练样本,通过自适应学习获得适合样本特征的冗余字典,它充分利用了字典原子与待重建图像的相关性,获得了待重建图像的理想完备稀疏表示,从而降低了采样率,提高了图像重建质量.最后,以全变分作为正则化条件,采用交替迭代算法求解稀疏优化问题.仿真结果表明,该算法可以在低采样率下重建出高质量的图像.

关键词：压缩感知自适应冗余字典稀疏表示图像重建全变分

在线全文

学校读者我要写书评

暂无评论

结合自适应字典学习的稀疏贝叶斯重构

西安电子科技大学学报 2016年第4期43卷 1-4,122页

作者：王勇乔倩倩杨笑宇徐文娟贾拯陈楚楚高全学西安电子科技大学电子工程学院陕西西安710071 西安电子科技大学通信工程学院陕西西安710071

贝叶斯压缩感知是一种基于统计分析的压缩感知算法,具有很好的鲁棒性,能够充分利用信息间的相关性,它的重构依赖于图像的稀疏性表达.针对贝叶斯压缩感知的深层次稀疏化问题,笔者结合自适应字典学习思想,提出一种冗余自适应字典表示的稀... 详细信息

贝叶斯压缩感知是一种基于统计分析的压缩感知算法,具有很好的鲁棒性,能够充分利用信息间的相关性,它的重构依赖于图像的稀疏性表达.针对贝叶斯压缩感知的深层次稀疏化问题,笔者结合自适应字典学习思想,提出一种冗余自适应字典表示的稀疏贝叶斯学习算法.该算法对图像进行局部分块,从待重建图像的迭代中间图像分块中学习字典,并以该字典作为图像的稀疏变换基,通过稀疏贝叶斯学习算法获得稀疏解.实验结果表明,基于自适应字典的贝叶斯学习算法能提高稀疏化,明显改善图像的重构质量.

关键词：稀疏贝叶斯学习自适应字典贝叶斯压缩感知