检索结果-内蒙古大学图书馆

Cybersecurity, Internet of Things and Soft Computing (CITSC), Asia-Europe Conference on

作者： Yanchun Liu Lishuang Zhao College of Information Science and Technology Bohai University Jinzhou China

ISBN: (数字)9798331504205

ISBN: (纸本)9798331504212

The exponential growth of data requires powerful analytical methods and processing tools, and deep learning demonstrates powerful analytical and predictive capabilities. CNN is excellent model in the field of deep learning, enabling powerful feature abstraction and image understanding that goes beyond the accuracy of recognition using the human eye. However, while the CNN model is constantly approaching the precision limit, the network depth and model parameters have multiplied, and the huge memory consumption and excessive storage occupation cannot be applied on many embedded and mobile devices. This paper constructs a compression algorithm model, including convolution layer, filter and pruning algorithm, and focuses on structured pruning and filter importance calculation. Using MNIST data set, the LeNet-5 model was trained by parameter pruning according to different pruning ratios, and the model compression was realized. The application of the research results in this paper can significantly reduce the redundant parameters in the network model, thus achieving the purpose of reducing the computing cost and storage space, and promoting the development of deep learning.

关键词： Deep learning Training Costs Accuracy Computational modeling Memory management filtering algorithms Mobile handsets Convolutional neural networks Compression algorithms

来源：评论

学校读者我要写书评

暂无评论

System Identification for Yaw Dynamics Prediction Model of an Autonomous Underwater Glider: Comparison of N4SID and ARX Models Using Preprocessed Data

System Identification for Yaw Dynamics Prediction Model of a...

引用

Advancement in Data Science, E-learning and Information Systems (ICADEIS), International Conference on

作者： Yansha Ramadhan Pratama Muhammad Hanif Hibatullah Bambang Riyanto Trilaksono Egi Hidayat Simon Siregar Yansen Dwiputra School of Electrical Engineering Bandung Institute of Technology Bandung Indonesia Faculty of Applied Science Telkom University Bandung Indonesia

ISBN: (数字)9798331513320

ISBN: (纸本)9798331513337

This paper aims to compare two algorithms of system identification between Auto-Regressive with eXogenous inputs (ARX) and Numerical Subspace State Space System Identification (N4SID). This comparison is based on the development of a Model Predictive Control (MPC) for heading control of an Autonomous Underwater Glider (AUG). The comparison was done by compared best-fit values of those models, made from data through a noise filtering process. The filtering process is using a Maximum Likelihood-based Kalman Filter (ML-based KF). ML is used to get the optimal parameters of Q and R value for K $F$ , thus that the optimal value for both parameters can be achieved. Both models have good accuracy performance on best-fit percentage and have low errors. Nevertheless, the ARX model demonstrates a bit advantage over N4SID, achieving the highest best-fit percentage than N4SID model. The ARX achieved the best fit percentage of 71.76 % when the na and nb tuned at 6 and 4, respectively with the errors achieved 3.127 for MSE and 1.768 for RMSE. Then the N4SID achieved the best fit percentage of 71.75% at the seventh-order tuning with MSE and RMSE are 3.1586 and 1.7725, respectively. This algorithm will be used as a prediction model for MPC, in future.

关键词： Heuristic algorithms Noise Predictive models filtering algorithms Prediction algorithms Data models System identification Kalman filters Tuning Overfitting

来源：评论

学校读者我要写书评

暂无评论

A System Design for 3D Visualization of Underwater Multi-Target Tracking Based on Unity 3D and GMPHD Filter

A System Design for 3D Visualization of Underwater Multi-Tar...

引用

International Conference on Electronics, Information and Communications (ICEIC)

作者： Seungho Kim Scott Uk-Jin Lee Department of Computer Science and Engineering Major in Bio-Artificial Intelligence Hanyang University Ansan Republic of Korea Department of Computer Science and Engineering Hanyang University Ansan Republic of Korea

ISBN: (数字)9798331510756

ISBN: (纸本)9798331510763

This paper proposes a design for a visualization system for underwater multi-target tracking based on the Gaussian Mixture Probability Hypothesis Density (GMPHD) filter using Unity 3D engine. Traditional analysis methods for underwater tracking algorithms rely primarily on two-dimensional graphs and numerical data, making it challenging to intuitively understand the complex behavior of tracking systems in three-dimensional underwater environments. Our proposed system design provides a framework for 3D visualization of target movements, sonar sensor detection areas, and filter estimation results, along with an interactive interface for parameter adjustment. The proposed architecture consists of four main components: simulation core, visualization module, environment simulation, and user interface, which would enable researchers to analyze algorithm performance under various scenarios effectively. Through real-time visualization and parameter adjustment capabilities, the system is designed to offer immediate feedback on how different parameters affect tracking performance. This architectural approach aims to provide researchers with a more intuitive understanding of the GMPHD filter's operation and facilitate the optimization of tracking algorithms for underwater environments. The proposed system design demonstrates the potential of 3D visualization tools in advancing underwater multi-target tracking research and algorithm development.

关键词： Three-dimensional displays Target tracking Data visualization Systems architecture filtering algorithms Information filters Real-time systems Underwater tracking System analysis and design Sonar detection

来源：评论

学校读者我要写书评

暂无评论

Robust Attitude Estimation With Quaternion Left-Invariant EKF and Noise Covariance Tuning

Robust Attitude Estimation With Quaternion Left-Invariant EK...

引用

International Conference on Computer and Automation Engineering, ICCAE

作者： Yash Pandey Rahul Bhattacharyya Yatindra Nath Singh Department of Electronics Engineering Harcourt Butler Technical University Kanpur India Department of Electrical Engineering Indian Institute of Technology Kanpur Kanpur India

ISBN: (数字)9798331533816

ISBN: (纸本)9798331533823

Accurate estimation of noise parameters is critical for optimal filter performance, especially in systems where true noise parameter values are unknown or time-varying. This article presents a quaternion left-invariant extended Kalman filter (LI-EKF) for attitude estimation, integrated with an adaptive noise covariance estimation algorithm. By employing an iterative expectation-maximization (EM) approach, the filter can effectively estimate both process and measurement noise covariances. Extensive simulations demonstrate the superiority of the proposed method in terms of attitude estimation accuracy and robustness to initial parameter misspecification. The adaptive LI-EKF's ability to adapt to time-varying noise characteristics makes it a promising solution for various applications requiring reliable attitude estimation, such as aerospace, robotics, and autonomous systems.

关键词： Accuracy filtering Quaternions Noise Estimation filtering algorithms filtering theory Kalman filters Time-varying systems Tuning

来源：评论

学校读者我要写书评

暂无评论

Robust Fractional-Order-Based Von Mises Subband Adaptive filtering for Feedback Cancellation in Hearing Aids

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2025年

作者： Devi, R. Vanitha Vasundhara Natl Inst Technol Warangal Dept Elect & Commun Engn Warangal 506004 India

The issue of acoustic feedback poses a consistent challenge in the context of hearing aids since it imposes restrictions on the attainable amplification levels and has the potential to significantly diminish the perceptual audio quality through the generation of whistling sounds. In recent studies, researchers have employed delay-less multiband structured band-wise acoustic feedback alleviation techniques for hearing aids. However, the resilience of this system in the existence of non-Gaussian noise has yet to be thoroughly addressed. To tackle this matter, a robust and new cost function has been introduced based on the modification of the von Mises distribution with a scaling parameter. In addition, fractional lower-order models provide resilience to heavy-tailed distributions, enhanced precision, heightened versatility and suitability for diverse data formats. In light of this viewpoint, a robust fractional order von Mises-based subband acoustic feedback compensation technique is introduced for hearing aids. The method's efficacy has been evaluated by computer simulations involving speech and music signals at various signal-to-noise ratio (SNR) levels. The evaluation results show 2-3 dB decrease in misalignment and 2-3 dB enhancement in added stable gain in contrast to prior methodologies. Moreover, perceptual evaluation of speech quality and hearing aid speech quality indexes have been improved by approximate to 2 % and approximate to 10 % respectively as employed with the proposed technique.

关键词： Noise Cost function Hearing aids Filter banks filtering Adaptive filters Acoustics Prediction algorithms Convergence filtering algorithms Hearing aid prediction error method fractional order adaptive filtering feedback compensation

来源：评论

学校读者我要写书评

暂无评论

Research on Personalized Recommendation algorithms Based on Collaborative filtering

Research on Personalized Recommendation Algorithms Based on ...

引用

Computer Engineering and Intelligent Control (ICCEIC) International Conference on

作者： Hangrui Cao Zhan Shi Hao Yang School of Computer Engineering Nanjing Institute of technology Nanjing China

ISBN: (数字)9798331507992

ISBN: (纸本)9798331508005

Traditional recommendation algorithms often suffer from issues such as data sparsity, lack of memory association, and prediction accuracy. This paper proposes a collaborative filtering recommendation algorithm that integrates memory and balance factors. When analyzing user interests and preferences, there are optimal design from multiple aspects such as memory factors and novelty of objects. This paper introduces a memory weight function to reflect the temporal characteristics of user interests, and incorporates a balance factor into user similarity calculation to reduce the impact of popular items on user similarity calculation, thus improving the accuracy of user similarity calculation. Experimental validation is conducted on the MovieLens dataset using MAE as the evaluation metric. The results demonstrate significant improvements in the experimental metrics, indicating enhanced recommendation accuracy and improved resilience to data sparsity.

关键词： Measurement Accuracy Collaborative filtering Clustering algorithms filtering algorithms Prediction algorithms Indexes Recommender systems Resilience Intelligent control

来源：评论

学校读者我要写书评

暂无评论

Inertial navigation attitude velocity and position algorithms using quaternion Scaled Unscented Kalman filtering

Inertial navigation attitude velocity and position algorithm...

引用

Annual Conference of Industrial Electronics Society

作者： Wassim Khoder Bienvenu Fassinut-Mombot Mohammed Benjelloun Laboratoire d'Analyse des Systèmes du Littoral Université du Littoral Côte d''Opale Calais France

In this paper, a scaled unscented Kalman filter (SUKF) based on the quaternion concept is designed for determination of the attitude, velocity and position parameters in inertial navigation system (INS) under large attitude error conditions. In this feedback filter, only bias effects are considered to be independent states and are used to compensate for navigation errors. To preserve the nonlinear nature of unit quaternion, the weighted mean computation for quaternions is derived in rotational space as a barycentric mean with renormalization and a multiplicative quaternion-error is used for predicted covariance computation of the quaternion because it represents the distance from the predicted mean quaternion. The updates are performed using quaternion multiplication which guarantees that quaternion normalization is maintained in the filter. Since the quaternion process noise increases the uncertainty in attitude orientation, modeling it as a vector part of quaternion is considered. Simulation and experimental results indicate a satisfactory performance of the newly developed model.

关键词： Inertial navigation filtering algorithms Quaternions Kalman filters Position measurement Accelerometers Algorithm design and analysis Uncertainty Velocity measurement Acceleration

来源：评论

学校读者我要写书评

暂无评论

Research on algorithms of digital filtering under fuzzy conditions

Research on algorithms of digital filtering under fuzzy cond...

引用

International Conference on Actual Problems of Electronic Instrument Engineering (APEIE)

作者： Yu.M. Veshkurtsev E.D. Bychkov D.A. Titov Omsk State Technical University Russia

In this paper, a new digital filtering algorithm is developed for estimating nonstationary signals contaminated with additive random noise. The results of simulation of a fuzzy digital filter are presented.

关键词： filtering algorithms Digital filters

来源：评论

学校读者我要写书评

暂无评论

P-Class Phasor Measurement Unit algorithms using adaptive filtering to enhance accuracy at off-nominal frequencies

P-Class Phasor Measurement Unit algorithms using adaptive fi...

引用

IEEE International Conference on Smart Measurements for Future Grids (SMFG)

作者： Andrew J. Roscoe Ibrahim F. Abdulhadi Graeme M. Burt Department of Electronic and Electrical Engineering University of Strathclyde Glasgow UK

While the present standard C.37.118-2005 for Phasor Measurement Units (PMUs) requires testing only at steady-state conditions, proposed new versions of the standard require much more stringent testing, involving frequency ramps and off-nominal frequency testing. This paper presents two new algorithms for “P Class” PMUs which enable performance at off-nominal frequencies to be retained at levels comparable to the performance for nominal frequency input. The performances of the algorithms are compared to the “Basic” Synchrophasor Estimation Model described in the new standard. The proposed algorithms show a much better performance than the “Basic” algorithm, particularly in the measurements of frequency and rate-of-change-of-frequency at off-nominal frequencies and in the presence of unbalance and harmonics.

关键词： Finite impulse response filter Frequency measurement filtering algorithms Phasor measurement units Correlation Frequency modulation IIR filters

来源：评论

学校读者我要写书评

暂无评论

RLS Adaptive filtering Without A Desired Signal: QR algorithms And Architectures

RLS Adaptive Filtering Without A Desired Signal: QR Algorith...

引用

Asilomar Conference on Signals, Systems & Computers

作者： P.S. Lewis Mechanical and Electronic Engineering Los Alamos National Laboratory Los Alamos NM USA

Most adaptive filters require a desired signal for operation. However, in many applications the a priori knowledge consists of the signal-to-data cross-correlation vector rather than a desired signal. Recursive sample matrix inversion (SMI) algorithms exist for this "mixed' case. These SMI algorithms, which are based on the inversion of a data cor- relation matrix, have both numerical and structural short- comings. This paper demonstrates how to formulate the re- cursive solution to the mixed case as a least squares problem. This formulation leads to algorithms based on recursive QR decomposition implemented by either Givens or fast Givens rotations. Compared to the recursive SMI approach, these QR-based algorithms are more efficient, have better numerical properties, and exhibit greater structural regularity. Because of their structural regularity, the algorithms are easily implemented by either a triangular or linear systolic array.

关键词： Resonance light scattering Adaptive filters filtering algorithms Matrix decomposition Vectors Autocorrelation Least squares approximation Recursive estimation Kalman filters Laboratories

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：