检索结果-内蒙古大学图书馆

5th International Symposium on Instrumentation and Control Technology

作者： Shi, ZK Northwestern Polytech Univ Xian 710072 Peoples R China

ISBN: (纸本)0819451371

A two-step robust filtering algorithm for uncertain discrete-time system is presented. To get a series of computational equations, the upper bound of uncertain parts generated by uncertain systematic matrix in the expression of the error covariance matrix of state estimation is given and the equivalent systematic matrix is obtained. These results build up a robust time update algorithm. On the other hand, the lower bound of uncertain parts generated by uncertain observation matrix in the expression of the error covariance matrix of state estimation is given and the equivalent observation matrix is obtained. Thus both the time update and measurement update algorithms are suggested. By means of matrix inversion formula, the expression structures of both time update and measurement update algorithms are all simplified.

关键词： robust estimation Kalman filter filtering algorithm optimal estimation

来源：评论

学校读者我要写书评

暂无评论

A new parameters adaptively adjusting method of current statistical model

A new parameters adaptively adjusting method of current stat...

引用

IEEE International Conference on Information and Automation 2015

作者： Yang, Yongjian Fan, Xiaoguagn Wang, Shengda Zhuo, Zhenfu Nan, Jianguo Huang, Lei Air Force Engn Univ Aeronaut & Astronaut Engn Coll Xian 710038 Shannxi Provinc Peoples R China

ISBN: (纸本)9781467391047

The fixed maximum acceleration and maneuvering frequency of current statistical model leads to the divergence of filtering algorithm. In this study, a new model which employs innovation dominated subjection function to adaptively adjust maximum acceleration and maneuvering frequency is proposed based on current statistical model. Although the new model has a better performance, a fluctuant phenomenon appears. As far as this problem is concerned, a new filter algorithm which is based on amendatory and adaptively fading kalman filtering is proposed. The results of simulation indicate the effectiveness and coherent of the new model and the new algorithm, and their well performance in maneuvering target tracking.

关键词： current statistical model filtering algorithm maneuvering target tracking

来源：评论

学校读者我要写书评

暂无评论

Near Field Distance Awareness algorithm using Bluetooth for Device Sociality Service 18

Near Field Distance Awareness Algorithm using Bluetooth for ...

引用

18th International Conference on Advanced Communication Technology (ICACT)

作者： Jung, Joon-young Kang, Dong-oh Choi, Jang-ho Min, Ok-gee Elect & Telecommun Res Inst Human Comp Res Sect Daejeon South Korea

ISBN: (纸本)9788996865063

The use of Device-to-Device (D2D) communication in various networks is expected to grow in the coming years. The D2D distance-based device sociality service is affected by distance awareness between the D2D devices. Thus, the wireless distance awareness between smart devices should be accomplished easily, accurately, and immediately. When smart devices are used in the distance-based distance sociality service, the smart device needs to know the distance to other smart devices in order to aware the space sociality. Several methods, such as receive signal strength indication (RSSI), time of arrival (ToA), and time difference of arrival (TDoA), can be used to aware the distance between D2D devices. Among these methods, the RSSI system can aware the D2D distance easily and inexpensively because most smart devices can estimate the received signal strength. However, estimating the distance using a RSSI is difficult due to inaccuracies. We conducted a preliminary test to discover the relationship between the actual distance and a Bluetooth RSSI under near field environment. Through the results of this test, we realize that the near field distance is hard to be classified due to the inaccuracy of Bluetooth RSSI. Therefore, in this paper, the near field distance awareness algorithm is proposed to reduce measurement errors by alleviating fluctuations in a Bluetooth signal. To evaluate the effectiveness of the proposed algorithm, the distance awareness is compared using different filtering algorithms, such as, a low-pass filer (LPF), a Kalman filter, and a particle filter under a meeting room environment. The proposed algorithm showed the best results in terms of the coefficient of determination, standard deviation, and measurement range.

关键词： Near field awareness Device to Device (D2D) Distance awareness filtering algorithm Wireless signal

来源：评论

学校读者我要写书评

暂无评论

Dynamic global constraints in backtracking based environments

引用

ANNALS OF OPERATIONS RESEARCH 2003年第1-4期118卷 101-119页

作者： Barták, R Charles Univ Prague Fac Math & Phys CR-18000 Prague Czech Republic

Global constraints provide strong filtering algorithms to reduce the search space when solving large combinatorial problems. In this paper we propose to make the global constraints dynamic, i.e., to allow extending the set of constrained variables during search. We describe a generic dynamisation technique for an arbitrary monotonic global constraint and we compare it with the semantic-based dynamisation for the all different constraint. At the end we sketch a dynamisation technique for non-monotonic global constraints. A comparison with existing methods to model dynamic problems is given as well.

关键词： global constraints filtering algorithm dynamic models

来源：评论

学校读者我要写书评

暂无评论

algorithm Design of Wearable Pedometer

Algorithm Design of Wearable Pedometer

引用

International Conference on Computer Systems, Electronics and Control (ICCSEC)

作者： Wei Yanxin Fan Xiujuan Yang Dongchen Beijing Inst Fash Technol Coll Informat Engn Beijing Peoples R China

ISBN: (纸本)9781538635735

With the rapid development of wearable devices, this paper improves the main step counting algorithm, and adds the functions of computing and displaying the data such as velocity, mileage and calorie consumption. The algorithm mainly includes the step algorithm of human motion data acquisition and gesture analysis, the filter algorithm based on Fourier decomposition, the filter algorithm based on mean value algorithm and the energy consumption algorithm for calories. These algorithms are written through the core processor. And single chip microcomputer will collect the calculated movement steps, mileage, current calories consumption and other motion data through Bluetooth components to the mobile phone, to achieve the monitoring of human movement. It has the characteristics of high precision, low redundancy and rich function.

关键词： step algorithm filtering algorithm energy consumption algorithm Bluetooth communication component

来源：评论

学校读者我要写书评

暂无评论

FILTER-ENHANCED HYPERGRAPH TRANSFORMER FOR MULTI-BEHAVIOR SEQUENTIAL RECOMMENDATION 49

FILTER-ENHANCED HYPERGRAPH TRANSFORMER FOR MULTI-BEHAVIOR SE...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Shao, Zhufeng Wang, Shoujin Lu, Wenpeng Zhang, Weiyu Guan, Hongjiao Zhao, Long Qilu Univ Technol Shandong Acad Sci Natl Supercomp Ctr JinanKey Lab Comp Power Netwo Shandong Comp Sci CtrMinist Educ Key Lab Comp Power Network & Informat Secur Jinan Peoples R China Univ Technol Sydney Data Sci Inst Sydney NSW Australia Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan Peoples R China

ISBN: (纸本)9798350344868;9798350344851

Sequential recommendation has been developed to predict the next item in which users are most interested by capturing user behavior patterns embedded in their historical interaction sequences. However, most existing methods appear to exhibit limitations in modeling fine-grained dependencies embedded in users' various periodic behavior patterns and heterogeneous dependencies across multi-behaviors. Towards this end, we propose a Filter-enhanced Hypergraph Transformer framework for Multi-Behavior Sequential Recommendation (FHT-MB) to address the above challenges. Specifically, a multi-scale filter layer equipped with multi-learnable filters is devised to encode behavior-aware sequential patterns emerging from different periodic trends (e.g., daily or weekly routines), and then a hypergraph structure is devised to extract heterogeneous dependencies across users' multiple types of behaviors. Extensive experiments on two real-world e-commerce datasets show the superiority of our proposed FHT-MB over various state-of-the-art methods. (1)

关键词： Multi-Behavior Sequential Recommendation filtering algorithm Hypergraph Learning

来源：评论

学校读者我要写书评

暂无评论

Inner Products Computation Using Folded Waveforms and Distributed Arithmetic Principles

Inner Products Computation Using Folded Waveforms and Distri...

引用

54th IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Pham, Duc-Minh Premkumar, A. B. Madhukumar, A. S. Nanyang Technol Univ Sch Comp Engn Singapore 639798 Singapore

ISBN: (纸本)9781612848570

In this paper, inner products using a new number system called the Folding Residue Number System (FRNS) and Distributed Arithmetic (DA) principles is proposed. The advantage of this new number system is that it allows digital filters to be implemented directly from the analog signal by decomposing and folding it in several parallel channels. Folding Residues extracted together with the folding information obtained from the folding circuits are processed using DA principles in implementing the filtering algorithm. The FRNS has the same complexity of computation as that of the Residue Number System (RNS) but intermediate steps such as conversion of the analog signal into its binary equivalents and subsequently into residues required in RNS are eliminated. Moreover, smaller word length of the folding residues coupled with DA results in faster implementation of the filtering algorithms.

关键词： Clocks DA principles FRNS Finite impulse response filter Table lookup analog signal binary equivalents distributed arithmetic distributed arithmetic principles filtering algorithm filtering theory folded waveforms folding residue number system inner products computation residue number systems waveform analysis

来源：评论

学校读者我要写书评

暂无评论

Learnable Filter Components for Social Recommendation 34th

Learnable Filter Components for Social Recommendation

引用

34th International Conference on Database and Expert Systems Applications (DEXA)

作者： Zhang, XianKun Huang, WenJie Univ Sci & Technol Sch Tianjin Tianjin Peoples R China

ISBN: (纸本)9783031398209;9783031398216

The recommendation system based on graph neural networks has gained increasing attention due to its superior learning ability of various additional information. Among them, the recommendation system based on social network graphs integrates social network graphs with user-item-graph interactions, aiming to extract dynamic preference features from recorded user behavior data to obtain more accurate recommendation results. However, recorded user behavior data contains noise and exhibits skewed distributions, which may lead to slightly insufficient representation performance of graph neural network models. This article proposes a new approach to address this issue by applying filtering algorithms commonly used in signal processing to noise reduction in the frequency domain. In this study, we combine learnable filter components with an MLP architecture to learn user and item latent embeddings in modeling social network graphs and user-item interaction graphs, and ultimately predict ratings using the latent embeddings of users and items. The entire MLP architecture has low time complexity, and the adaptive filter components are able to effectively suppress noise information. The model proposed in this article has been applied to real datasets and outperformed other models, achieving a 3.2% increase in the MAE evaluation index.

关键词： Social Recommendation Graph Neural Network filtering algorithm

来源：评论

学校读者我要写书评

暂无评论

Continuous-time Nonlinear Estimation Filters Using UKF-aided Gaussian Sum Representations

Continuous-time Nonlinear Estimation Filters Using UKF-aided...

引用

16th International Conference on Information Fusion (FUSION)

作者： Gokce, Murat Kuzuoglu, Mustafa Sci & Technol Res Council Turkey Technol & Innovat Funding Programs Directorate Ankara Turkey Middle East Tech Univ Dept Elect & Elect Engn Ankara Turkey

ISBN: (纸本)9786058631113

An approximate nonlinear estimation method for continuous-time systems with discrete-time measurements is developed. The approach evaluates the Gaussian sum approximation of the a priori probability density function (pdf) by solving the Fokker-Planck equation numerically. Approximate evaluation of the a posteriori pdf is achieved by using Gaussian sums, a priori pdf and measurements in Bayes rule. Mean and covariance values of Gaussians are chosen by the help of an Unscented Kalman Filter (UKF), with respect to a region where a priori and a posteriori pdfs are approximated. Weights of the Gaussians are updated using the deterministically chosen grid points in the specified domains. UKF here acts as a one step look ahead mechanism to determine the high probability regions where a priori and a posteriori pdfs can reside. The a priori and a posteriori pdfs are approximated around these high probability regions. The developed approach is compared with UKF and Particle Filter in a one dimensional nonlinear system.

关键词： nonlinear system filtering algorithm continuous-time system Fokker-Planck equation Gaussian sum numerical methods

来源：评论

学校读者我要写书评

暂无评论

Location & Tracking of Bistatic System in Jamming

Location & Tracking of Bistatic System in Jamming

引用

1996 CIE International Conference of Radar(ICR’96)

作者： Chen Yongguang Sun Zhongkang Institute of Electronic Engineering National University of Defense Technology

ISBN: (纸本)7505339036

Under the condition of januning the number of usable measurements in a T/R-R bistatic system is *** it is necessary to study some location and tracking algorithms concerning with different sets of measurement *** multi-stage tracking teclmique proposed in the paper makes full use of measurement data in T/R and R stations under different condition of jamming,as a result the tracking precision of 3-D moving targets is *** performance of these algorithms is evaluated with the help of computer simulation for typical target paths.

关键词： Multi-stage tracking Bistatic system filtering algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：