检索结果-内蒙古大学图书馆

31st ACM Web Conference (WWW)

作者： Zhou, Kun Yu, Hui Zhao, Wayne Xin Wen, Ji-Rong Renmin Univ China Sch Informat Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Renmin Univ China Gaoling Sch Artificial Intelligence Beijing Peoples R China Beijing Key Lab Big Data Management & Anal Method Beijing Peoples R China Chinese Acad Sci Inst Geol & Geophys Key Lab Petr Resources Res Beijing Peoples R China

ISBN: (纸本)9781450390965

Recently, deep neural networks such as RNN, CNN and Transformer have been applied in the task of sequential recommendation, which aims to capture the dynamic preference characteristics from logged user behavior data for accurate recommendation. However, in online platforms, logged user behavior data is inevitable to contain noise, and deep recommendation models are easy to overfit on these logged data. To tackle this problem, we borrow the idea of filtering algorithms from signal processing that attenuates the noise in the frequency domain. In our empirical experiments, we find that filtering algorithms can substantially improve representative sequential recommendation models, and integrating simple filtering algorithms ( e.g., Band-Stop Filter) with an all-MLP architecture can even outperform competitive Transformer-based models. Motivated by it, we propose FMLP-Rec, an all-MLP model with learnable filters for sequential recommendation task. The all-MLP architecture endows our model with lower time complexity, and the learnable filters can adaptively attenuate the noise information in the frequency domain. Extensive experiments conducted on eight real-world datasets demonstrate the superiority of our proposed method over competitive RNN, CNN, GNN and Transformerbased methods. Our code and data are publicly available at the link: https://***/RUCAIBox/FMLP-Rec.

关键词： Sequential Recommendation All-MLP Model filtering algorithm

来源：评论

学校读者我要写书评

暂无评论

ESTIMATION OF WIDELY FACTORIZABLE HYPERCOMPLEX SIGNALS WITH UNCERTAIN OBSERVATIONS 44

ESTIMATION OF WIDELY FACTORIZABLE HYPERCOMPLEX SIGNALS WITH ...

引用

44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Fernandez-Alcala, R. M. Jimenez-Lopez, J. D. Navarro-Moreno, J. Ruiz-Molina, J. C. Le Bihan, N. Univ Jaen Dept Stat & Operat Res Campus Las Lagunillas Jaen 23071 Spain Univ Grenoble Alpes CNRS Grenoble INP Gipsa Lab F-38000 Grenoble France

ISBN: (纸本)9781479981311

The filtering estimation problem under uncertainty conditions is addressed for a class of improper quaternion signals, called widely factorizable, characterized because their augmented correlation function is a factorizable kernel. From the knowledge of the correlation functions involved, a recursive algorithm is designed for the computation of the widely linear (WL) filtering estimate and its associated mean squared error. The main advantage of the proposed solution is that it can be applied in situations where a state-space model is not readily at hand. The benefits of the proposed WL filtering algorithm is analyzed through a simulation example where WL filtering errors are compared with respect to the strictly linear (SL) counterparts, showing the superior behavior of the former over the latter.

关键词： Quaternion signals filtering algorithm uncertainty conditions widely factorizable signals widely linear processing

来源：评论

学校读者我要写书评

暂无评论

Research on the Detection Model of Wearable Fall Detection algorithm

Research on the Detection Model of Wearable Fall Detection A...

引用

2023 International Conference on Integrated Intelligence and Communication Systems, ICIICS 2023

作者： Yao, Fenggang Kang, Yuru Wei, Yanxi School of Artificial Intelligence Xi'an Aeronautical Polytechnic Institute Xi'an China

ISBN: (纸本)9798350315455

Falls are very common today and and this behavior can lead to serious injuries and even accidental death in the elderly when they fall, especially among the elderly.. In order to avoid the injuries caused by the elderly who are not able to get timely assistance after a fall, this paper investigates fall detection algorithms that can reduce the injury and impact of falls on the elderly. In this paper, A bivearnerable fall detection algorithm based on a bidirectional long and short-term memory neural network (LSTM) will be presented to improve the accuracy and real-time performance of the detection algorithm. The algorithm automatically extracts the fall behavior data from the input data by taking the data from the inertial sensors, and the deeper data features from inside for the data to be pre-processed and processed to the detection results. The features from the acceleration sensors are extracted in the algorithm model and used to detect the algorithm model. In this paper, the algorithm model is validated with the SisFall dataset and the result shows bidirectional LSTM obtained 95.30% accuracy, which has good practicality and strong generalization ability. © 2023 IEEE.

关键词： bidirectional LSTM dimensionality reduction features filtering algorithm neural network wearable computer

来源：评论

学校读者我要写书评

暂无评论

Optimized Robust Filter for Uncertain Discrete Time System and Its Application to Flight Test

引用

Chinese Journal of Aeronautics 2003年第2期16卷 91-96页

作者：史忠科 Department of Automatic Control Northwestern Polytechnical University

An optimized robust filtering algorithm for uncertain discrete-time systemsis presented. To get a series of computational equations, the uncertain part generated by theuncertain systematic matrix in the expression of the error-covariance matrix of time update stateestimation is optimized and the least upper bound of the uncertain part is given. By means of theseresults, the equivalent systematic matrix is obtained and a robust time update algorithm is builtup. On the other hand, uncertain parts generated by the uncertain observation matrix in theexpression of the error-covariance matrix of measurement update state estimation are optimized, andthe largest lower bound of the uncertain part is given. Thus both the time update and measurementupdate algorithms are developed. By means of the matrix inversion formula, the expression structuresof both time update and measurement update algorithms are all simplified. Moreover, the convergencecondition of a robust filter is developed to make the results easy to application. The results offlight data processing show that the method presented in this paper is efficient.

关键词： robust estimation Kalman filter filtering algorithm optimal estimation flight test

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Circuit Breaker with synchronous Closing Function Based on DSP

An Intelligent Circuit Breaker with synchronous Closing Func...

引用

30th Annual Conference of the IEEE-Industrial-Electronics-Society (IECON)

作者： Cai Zhyuan Yu Li Ma Shaohua Shenyang Univ Technol Coll Elect Engn Shenyang Peoples R China

ISBN: (纸本)9780780387300

An intelligent circuit breaker with synchronous closing function based on DSP is presented in this paper. By which, the filtering algorithm based on Fourier algorithm is adopted in order to estimate theoptimal reclosing angle so as to weaken the instantaneous current during reclosing circuit. Based on the discussion of the digital signal processing technology, the hardware circuit has been set up and the software has been programmed. And the simulation results show a good agreement with the predictions.

关键词： circuit breakers Fourier analysis short-circuit currents power engineering computing digital signal processing chips intelligent circuit breaker synchronous closing function DSP filtering algorithm Fourier algorithm optimal reclosing angle digital signal processing technology hardware circuit

来源：评论

学校读者我要写书评

暂无评论

A study on multi-stage tracking technique in T/R-R bistatic system

A study on multi-stage tracking technique in T/R-R bistatic ...

引用

IEEE 1995 National Aerospace and Electronics Conference (NAECON 1995)

作者： Chen, YG Sun, ZK Cheng, HW NATL UNIV DEF TECHNOL DEPT ELECTR TECHCHANGSHA 410073HUNANPEOPLES R CHINA

ISBN: (纸本)0780326679

When the communication link of a T/R-R bistatic radar system is affected by external interference, the number of usable measurements in it is variable. Therefore it is necessary to study some tracking algorithms concerning with different sets of measurement data. The multi-stage tracking technique proposed in the paper makes full use of measurement data in T/R and R station under different condition, as a result the tracking precision of 3-D moving targets is increased. The performance of these algorithms is evaluated with the help of computer simulation for typical target paths.

关键词： multi-stage tracking bistatic radar system filtering algorithm

来源：评论

学校读者我要写书评

暂无评论

Multi-scenario surface temperature estimation in lithium-ion batteries with transfer learning and LGT augmentation

引用

ENERGY 2024年 304卷

作者： You, Yuqiang Lin, Mingqiang Meng, Jinhao Wu, Ji Wang, Wei Chinese Acad Sci Quanzhou Inst Equipment Mfg Fujian Inst Res Struct Matter Jinjiang 362200 Fujian Peoples R China Univ Chinese Acad Sci Fujian Coll Fuzhou 350002 Peoples R China Fujian Key Lab Special Intelligent Equipment Safet Fuzhou 350003 Peoples R China Xi An Jiao Tong Univ Sch Elect Engn Xian 710049 Peoples R China Hefei Univ Technol Sch Automot & Transportat Engn Hefei 230009 Peoples R China Xi An Jiao Tong Univ Sch Management Xian 710049 Peoples R China

The State of Temperature (SOT) plays a crucial role in ensuring the safety and reliability of lithium-ion batteries, as well as the stability of electric vehicles (Evs). Recently, data-driven methods for lithium-ion battery temperature estimation have often obtained short-term estimation information in a single scenario. To alleviate the above issues, this paper proposes a method for estimating the surface temperature of lithium-ion batteries based on Local and Global Trend (LGT) data augmentation and Long Short-Term Memory (LSTM) model transfer. Initially, voltage and current data are processed using two types of filtering methods to reduce noise while preserving the fluctuating characteristics of the original data. Differential features are subsequently extracted from the polynomially filtered curves. Then all the data, including the filtered data and the features are fed into the LGT data augmentation algorithm, followed by the use of an LSTM transferable model to estimate the surface temperature of the target battery. Furthermore, the method is tested on both laboratory datasets and datasets with real driving conditions, covering a wide range of environmental temperatures and driving scenarios typical of EVs. The experimental results demonstrate that retraining with only the first 30 % of the target battery's data yields effective SOT estimation results.

关键词： Surface temperature filtering algorithm Data enhancement Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Updating numerical models towards time domain alignment of structural dynamic responses with a limited number of sensors

引用

MECHANICAL SYSTEMS AND SIGNAL PROCESSING 2023年 204卷

作者： Fu, Yuhao Wang, Ying Harbin Inst Technol Shenzhen Sch Civil & Environm Engn Shenzhen 518055 Guangdong Peoples R China Guangdong Prov Key Lab Intelligent & Resilient Str Shenzhen Guangdong Peoples R China

Due to the inherent uncertainty and complexity of civil infrastructure, it is challenging to simulate their structural dynamic responses accurately and efficiently. Existing model updating methods normally need distributed sensors and use modal parameters to construct objective functions, which can achieve frequency domain agreement between the numerical model and the real structure. Nevertheless, time domain alignment is still difficult to achieve, especially when the sensor number is limited. To enhance the simulation accuracy, this paper proposes a two-step model updating approach through updating in frequency and time domains sequentially. A virtual force is introduced in the second step to represent structural complexity and input error, and is derived by using three representative algorithms. The approach was verified on a scaled wind turbine model in the laboratory. Parametric studies were performed on different arrangements of sensors and virtual force. The results show that the virtual force should be placed at a sensor point and close to the acting point of external loads. The approach can achieve time domain alignment between numerical model and real structure with a normalized mean square error averaging 3.25% in the optimal setting. Based on the results, the proposed framework may find vast application in the establishment of numerical models towards structural dynamic twins.

关键词： Virtual force Force identification System identification filtering algorithm Structural health monitoring

来源：评论

学校读者我要写书评

暂无评论

Angle-Weighted trilateration method fusing multi-data processing technologies in indoor scene

引用

DIGITAL SIGNAL PROCESSING 2024年 147卷

作者： Zang, Liguo Jiao, Jing Wang, Jie Ding, Ning Wang, Zizhou Min, Xinyi Nanjing Inst Technol Sch Automobile & Rail Transportat Nanjing 211167 Peoples R China

The trilateration method based on Ultra-Wideband (UWB) technology is widely used in indoor positioning. However, the complexity of indoor environment will affect the transmission of UWB signals, reducing accuracy of trilateration method. Therefore, an Angle -Weighted trilateration method fusing multi -data processing technologies is utilized. By analyzing the relationship between positioning accuracy and inner angles of the error triangle generated by trilateration method, the rule of measuring positioning accuracy with angle as weight is determined. The triangle is corrected by the rule, and the corrected triangle is solved by centroid method to obtain accurate positioning results. In addition, the linear fitting method, Kalman Filter algorithm and Unscented Kalman Filter algorithm are used to process ranging and positioning results of the system, aiming to reduce error range of positioning. In static environment, the Root Mean Square Error of the utilized method is about 10 mm, and the Mean Absolute Error is within 3 mm. In dynamic environment, the Root Mean Square Error of the utilized method is within 40 mm, and the Mean Absolute Error is within 20 mm. The Angle -Weighted trilateration method fusing multi -data processing technologies can improve accuracy and stability of the positioning system.

关键词： Angle -Weighted trilateration filtering algorithm Indoor positioning Linear fitting method UWB

来源：评论

学校读者我要写书评

暂无评论

Simultaneous input and state estimation for stochastic nonlinear systems with additive unknown inputs

引用

AUTOMATICA 2020年 111卷 108588-000页

作者： Kim, Hunmin Guo, Pinyao Zhu, Minghui Liu, Peng Univ Illinois Dept Mech Sci & Engn 1206 W Green St Urbana IL 61801 USA Airbnb Inc 888 Brannan St San Francisco CA 94103 USA Penn State Univ Sch Elect Engn & Comp Sci University Pk PA 16802 USA Penn State Univ Coll Informat Sci & Technol 201 Old Main University Pk PA 16802 USA

This paper investigates simultaneous input and state estimation for a class of nonlinear stochastic systems. We propose a recursive filter to concurrently estimate system states and unknown inputs. We show that the estimation errors of the proposed filter are Practically Exponentially Stable in probability, and the estimation error covariance matrices are uniformly bounded. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： Nonlinear systems Stochastic systems State estimation Input estimation 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：