检索结果-内蒙古大学图书馆

Advances in Electrical, Electronics and Computing Technology (EECT), International Conference on

作者： Jinjian Jiang Chenzhan Zeng Yong Wang Department of Automation University of Science and Technology of China Hefei China

ISBN: (数字)9798331541545

ISBN: (纸本)9798331541552

This paper is the first to investigate the state filtering problem for linear fractional order discrete state-space systems with state constraints. To tackle this issue, a reduced-order fractional system model is presented by utilizing the linear relationships between state variables and constrains. Afterwards, a constrained reduced-order fractional Kalman filter(CRFKF) is derived accordingly. Last, the effectiveness of the proposed algorithm is demonstrated by a simple tracking example.

关键词： Linear systems Computers filtering Stochastic systems Simulation Computational modeling filtering algorithms Kalman filters State estimation Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

LINE SEARCH algorithms FOR ADAPTIVE filtering

引用

IEEE TRANSACTIONS ON SIGNAL PROCESSING 1993年第7期41卷 2490-2494页

作者： DAVILA, CE Department of Electrical Engineering Southern Methodist University Dallas TX USA

Adaptive filtering algorithms which adjust their step size in order to minimize the sum of squared estimation errors on a linear manifold or ''line'' are described. The resulting algorithms have least mean square (LMS) complexity, zero asymptotic misadjustment for the stationary data case, and converge faster than LMS.

关键词： filtering algorithms Adaptive filters Convergence Adaptive equalizers Least squares approximation Nonlinear filters Vectors Error correction Covariance matrix Stochastic processes

来源：评论

学校读者我要写书评

暂无评论

Recurrent Difference Digital filtering Algorithm Without Algorithmic Multiplication Operations

Recurrent Difference Digital Filtering Algorithm Without Alg...

引用

Digital Signal Processing and its Applications (DSPA), International Conference on

作者： Adeliya Yu. Burova Moscow Aviation Institute (National Research University) Moscow Russia

ISBN: (数字)9798331518752

ISBN: (纸本)9798331518769

Research-related issues of any possibility on building virtual models for microelectronic digital signal processing devices without multiplication operations were considered. The research relevance was due to the need of some reducing of the computational complexity of software algorithms for digital filtration. The research purpose was to develop a mathematical model of the recursive digital filter with low-bit values for its coefficients of digital filtering based on the finite difference method. The study used numerical methods for constructing mathematical and software models of digital algorithms for difference filtering with such coefficients. The research results confirmed and showed the real possibility for building of mathematical model for digital algorithms of these filters by the methods for directional search of integer difference coefficients for difference filtering and comparative analysis of such low-bit difference coefficients. Recurrent formula for linear convolution of sequences of digital samples of the filtered signal and difference coefficients of the recursive difference digital filter is given.

关键词： Convolution Computational modeling Software algorithms Signal processing algorithms Nonlinear filters filtering algorithms Mathematical models Software Numerical models Digital filters

来源：评论

学校读者我要写书评

暂无评论

Speech Data Selection for Efficient ASR Fine-Tuning using Domain Classifier and Pseudo-Label filtering

Speech Data Selection for Efficient ASR Fine-Tuning using Do...

引用

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

作者： Pradeep Rangappa Juan Zuluaga-Gomez Srikanth Madikeri Andres Carofilis Jeena Prakash Sergio Burdisso Shashi Kumar Esaú Villatoro-Tello Iuliia Nigmatulina Petr Motlicek Karthik Pandia Aravind Ganapathiraju Idiap Research Institute Martigny Switzerland EPFL Lausanne Switzerland Uniphore Software Systems India University of Zurich Switzerland Brno University of Technology Czech Republic

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In real-world speech data processing, the scarcity of annotated data and the abundance of unlabelled speech data present a significant challenge. To address this, we propose an efficient data selection pipeline for fine-tuning ASR models by generating pseudo-labels using WhisperX pipeline and selecting efficient labels for fine-tuning. In our work, we propose a domain classifier system developed with a computationally inexpensive TFIDF and classical machine learning algorithm. Later, we filter data from the classifier output using a novel metric that assesses word ratio and perplexity distribution. The filtered pseudo labels are then used for fine-tuning standard encoder-decoder Whisper models and Zipformer. Our proposed data selection pipeline reduces the dataset size by approximately 1/100 th while maintaining performance comparable to the full dataset, outperforming random domain-independent selection strategies.

关键词： Measurement Training Adaptation models Machine learning algorithms filtering Computational modeling Pipelines filtering algorithms Data models Speech processing

来源：评论

学校读者我要写书评

暂无评论

COMPARATIVE STUDY OF 3-D IMAGE RECONSTRUCTION algorithms WITH REFERENCE TO NUMBER OF PROJECTIONS AND NOISE filtering

引用

IEEE TRANSACTIONS ON NUCLEAR SCIENCE 1975年第1期NS22卷 344-358页

作者： CHO, ZH CHAN, JK HALL, EL KRUGER, RP MCCAUGHEY, DG UNIV CALIF ELECT SCI & ENGR DEPTLAB NUCL MED & RADIOL BIOLLOS ANGELESCA 90024 UNIV SO CALIF IMAGE PROCESSING INSTELECT ENGR DEPTLOS ANGELESCA

A comparative study of four popular 3-D image reconstruction algorithms has been made. Particular attention was given to artifacts generated and noise sensitivity. The methods considered include two spatial domain convolution algorithms, the Linear Superposition with Compensation (LSC) and a Fourier Convolution Method (FCM), a direct Fast Fourier Transform method (FFT), and an algebraic technique, the Simultaneous Iterative Reconstruction Technique (SIRT). The methods were compared by computing reconstructed images for an identical input phantom image. The phantom image contains several edges and a 2% contrast object. Variations, artifacts and noise sensitivity are easily visualized by perspective plots of the reconstructed images. Considerations as to the optimum algorithm for a particular application are discussed.

关键词： Image reconstruction filtering algorithms Iterative algorithms Convolution Imaging phantoms Noise generators Fast Fourier transforms Iterative methods Quantum computing Visualization

来源：评论

学校读者我要写书评

暂无评论

An adaptive Kalman filtering algorithm for ship speed measurement

An adaptive Kalman filtering algorithm for ship speed measur...

引用

IEEE Information Technology and Mechatronics Engineering Conference (ITOEC)

作者： Shukui Zhang Qibing Dai Ding Xiong Navigation College Jiangsu Maritime Institute Nanjing China Digital Engineering Technology Research and Development Center for Maritime Safety and Security Nanjing China

ISBN: (数字)9798331529482

ISBN: (纸本)9798331529499

In response to the problem of low measurement accuracy caused by dynamic model errors in ship speed measurement using Kalman filtering, this article proposes an adaptive Kalman filtering algorithm: an abnormal error identification function is constructed to detect dynamic model errors, and an exponential function is used to construct an adaptive factor to reduce dynamic model errors. Simulation verification shows that compared with KF algorithm and interactive multiple model algorithm (IMM), AKF algorithm has reduced the mean absolute error (MAE) by 22% and 11% respectively, and the root mean square error (RSME) by 26% and 15% respectively, indicating that AKF algorithm has better performance in both measurement accuracy and stability evaluation indicators.

关键词： Adaptation models Accuracy filtering Heuristic algorithms Measurement uncertainty filtering algorithms Stability analysis Velocity measurement Kalman filters Marine vehicles

来源：评论

学校读者我要写书评

暂无评论

Distributionally Robust Kalman filtering over an Infinite-Horizon

Distributionally Robust Kalman Filtering over an Infinite-Ho...

引用

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

作者： Joudi Hajar Taylan Kargin Vikrant Malik Babak Hassibi Caltech Pasadena CA USA

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

This paper investigates distributionally robust filtering for state-space models subject to exogenous disturbances in state evolution and observation processes. The joint probability distribution of the disturbance process over an arbitrary horizon is unknown but is assumed to reside within a Wasserstein-2 ambiguity set centered around a nominal distribution. We seek to develop a causal estimator that minimizes the worst-case mean squared error (MSE) among all distributions in this ambiguity set. Unlike prior works, our framework accommodates disturbances with arbitrary temporal correlations. In the finite-horizon setting, this problem reduces to a semi-definite program (SDP) whose complexity scales with the time horizon. Consequently, we focus on the infinite-horizon case, where we derive the optimal linear time-invariant (LTI) filter using the Karush-Kuhn-Tucker (KKT) conditions and propose an efficient frequency-domain algorithm to compute it. While the optimal LTI filter generally has a non-rational transfer function, we provide a minimax rational approximation method to approximate the optimal non-rational filter, in the H ∞ -norm, with a near-optimal finite-order state-space estimator. This approach avoids the computational challenges associated with horizon-dependent scaling in the finite-horizon case. Numerical simulations demonstrate the effectiveness of the proposed method.

关键词： Linear systems filtering Transfer functions Signal processing algorithms filtering algorithms Signal processing Robustness Probability distribution State-space methods Speech processing

来源：评论

学校读者我要写书评

暂无评论

Efficiency Evaluation of a Combining Compact 1D CNN and Signal filtering Prototype Algorithm for Vibration Signal Classification

Efficiency Evaluation of a Combining Compact 1D CNN and Sign...

引用

First International Symposium on Instrumentation, Control, Artificial Intelligence, and Robotics (ICA-SYMP)

作者： Thosapon Krinsukont Taweedej Sirithanapipat National Electronics and Computer Technology Center National Science and Technology Development Agency Pathumthani Thailand Department of Mechanical Engineering Kasetsart University Bangkok Thailand

ISBN: (数字)9798331505400

ISBN: (纸本)9798331505417

This paper presents an experiment and results of the modified CNN algorithm, it was developed by combining a compact 1D convolution neural network with a tuned signal filter (low-pass filter in this experiment). The algorithm was specifically designed for signal shape detection, prediction, and classification. An experiment aimed to evaluate the created algorithm's structure and signal filter to ensure its suitability for vibration signal data analysis. A testbed was built using an inverter motor equipped with various unbalanced conditions by increasing screws. Four-class unbalanced conditions were studied compared with the normal condition. Each conditions produced the distinct signal shape, with slight differences, making them suitable for evaluating the proposed algorithm's performance. The results demonstrate the satisfied performance of the algorithm, which effectively classifies vibration signals unbalance-3 (precision 54%), unbalance-4 (precision 77 % ) and a group of normal, unbalance-1, and unbalance-2. The group of normal, unbalance-1, and unbalance-2 can't be easily classified due to the tiny differences in the signal shapes, caused by small mass variations of the unbalanced screw unbalance-1, and unbalance-2. In conclusion, the algorithm proved satisfied effectiveness for its intended purpose of vibration signal classification.

关键词： Vibrations Shape Convolution Neural networks Low-pass filters Pattern classification filtering algorithms Fasteners Prediction algorithms Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

A Novel Hybrid Adaptive filtering Algorithm Controlled STATCOM for LVRT Capability Enhancement of DFIG

A Novel Hybrid Adaptive Filtering Algorithm Controlled STATC...

引用

Smart and Sustainable Developments in Electrical Engineering (SSDEE), IEEE International Conference on

作者： Arindam Chakraborty Tanmoy Maity Nilanjan Senroy Department of Electrical Engineering Indian Institute of Technology Delhi New Delhi India Department of Electrical Engineering Indian Institute of Technology (ISM) Dhanbad Dhanbad Jharkhand India

ISBN: (数字)9798331542108

ISBN: (纸本)9798331542115

This work focuses on enhancing transient stability or low voltage ride-through (LVRT) in a doubly fed induction generator (DFIG) by dynamically tuning all the PI controllers of the STATCOM using a novel approach. The hybrid enhanced block sphere Bayesian’ algorithm (EBSBA) and ‘least mean square root of exponential’ (LMSRE) algorithms (EBSBA- LMSRE) is proposed as a solution, offering key advantages such as high convergence speed, low computational complexity, and a straightforward design. The main goal of this work is to address voltage fluctuations and restore stable conditions in the DFIG. In this work, the cascaded structure-based proportional integral (PI) controllers are implemented on the voltage source converter (VSC) of the static synchronous compensator (STATCOM). During the transient, the gain of the PI controllers is automatically adjusted by the proposed hybrid EBSBA-LMSRE algorithm. The EBSBA-LMSRE-PI controlled STATCOM demonstrates superior response and damping compared to the LMSRE-PI and EBESBA-PI controllers, especially under normal and abnormal grid disturbances.

关键词： Damping PI control Voltage fluctuations Fluctuations Heuristic algorithms Doubly fed induction generators Power system stability filtering algorithms Automatic voltage control Transient analysis

来源：评论

学校读者我要写书评

暂无评论

PARALLEL IMPLEMENTATION OF EFFICIENT LS algorithms FOR filtering AND PREDICTION

引用

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING 1987年第11期35卷 1565-1569页

作者： KALOUPTSIDIS, N THEODORIDIS, S UNIV PATRAS DEPT COMP ENGNPATRASGREECE

A highly concurrent algorithm for LS FIR filtering and prediction is derived in this paper. The resulting reflection coefficients associated with the solution of Toeplitz and near to Toeplitz systems of equations are computed in an order recursive manner. This results in a highly parallel algorithm requiring O(p) computing time and O(p) processors, p being the order of the filter.

关键词： filtering algorithms Finite impulse response filter Statistics Signal processing algorithms Equations Concurrent computing Symmetric matrices Reflection Parallel algorithms Nonlinear filters

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：