检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Qi, Jun Hu, Hu Wang, Yannan Yang, Chao-Han Huck Siniscalchi, Sabato Marco Lee, Chin-Hui Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States Computer Engineering School University of Enna Italy Tencent Media Lab Tencent Corporation ShenzhenGuangdong China

This paper investigates different trade-offs between the number of model parameters and enhanced speech qualities by employing several deep tensor-to-vector regression models for speech enhancement. We find that a hybrid architecture, namely CNN-TT, is capable of maintaining a good quality performance with a reduced model parameter size. CNN-TT is composed of several convolutional layers at the bottom for feature extraction to improve speech quality and a tensor-train (TT) output layer on the top to reduce model parameters. We first derive a new upper bound on the generalization power of the convolutional neural network (CNN) based vector-to-vector regression models. Then, we provide experimental evidence on the Edinburgh noisy speech corpus to demonstrate that, in single-channel speech enhancement, CNN outperforms DNN at the expense of a small increment of model sizes. Besides, CNN-TT slightly outperforms the CNN counterpart by utilizing only 32% of the CNN model parameters. Besides, further performance improvement can be attained if the number of CNN-TT parameters is increased to 44% of the CNN model size. Finally, our experiments of multi-channel speech enhancement on a simulated noisy WSJ0 corpus demonstrate that our proposed hybrid CNN-TT architecture achieves better results than both DNN and CNN models in terms of better-enhanced speech qualities and smaller parameter sizes. Copyright © 2020, The Authors. All rights reserved.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Tensor-to-vector regression for multi-channel speech enhancement based on tensor-train network

arXiv

引用

arXiv 2020年

We propose a tensor-to-vector regression approach to multi-channel speech enhancement in order to address the issue of input size explosion and hidden-layer size expansion. The key idea is to cast the conventional deep neural network (DNN) based vector-to-vector regression formulation under a tensor-train network (TTN) framework. TTN is a recently emerged solution for compact representation of deep models with fully connected hidden layers. Thus TTN maintains DNN's expressive power yet involves a much smaller amount of trainable parameters. Furthermore, TTN can handle a multi-dimensional tensor input by design, which exactly matches the desired setting in multi-channel speech enhancement. We first provide a theoretical extension from DNN to TTN based regression. Next, we show that TTN can attain speech enhancement quality comparable with that for DNN but with much fewer parameters, e.g., a reduction from 27 million to only 5 million parameters is observed in a single-channel scenario. TTN also improves PESQ over DNN from 2.86 to 2.96 by slightly increasing the number of trainable parameters. Finally, in 8-channel conditions, a PESQ of 3.12 is achieved using 20 million parameters for TTN, whereas a DNN with 68 million parameters can only attain a PESQ of 3.06. Code is available online. Copyright © 2020, The Authors. All rights reserved.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Power Efficient FPGA-based TCAM Architecture by using Segmented Matchline Strategy

Power Efficient FPGA-based TCAM Architecture by using Segmen...

引用

Advances in the Emerging Computing Technologies (AECT), International Conference on

作者： Najib Ur Rehman Omer Mujahid Zahid Ullah Abdul Hafeez Tama Fouzder Muhammad Ibrahim Department of Electrical Engineering CECOS University of IT & Emerging Sciences Peshawar Pakistan Department of Computer Science & IT University of Engineering and Technology Pakistan Department of Electrical and Electronic Engineering University of Liberal Arts Bangladesh Bangladesh

ISBN: (数字)9781728144528

ISBN: (纸本)9781728144535

Ternary content-addressable memory (TCAM) is famous for its high-speed search operation but this speed comes at the cost of high-power consumption. The paper presents a power efficient architecture for TCAM on field-programmable gate array (FPGA) by using segmented matchline (ML) strategy. Each ML is divided into four equal segments where each segment has nine bits. The proposed TCAM architecture is implemented on Xilinx Virtex-6 FPGA for the size of 64 × 36. Implementation results show that during search operation, the proposed TCAM results in 13.88% reduction in dynamic power consumption compared to the latest FPGA-based TCAMs while not compromising on speed.

关键词： Random access memory Field programmable gate arrays Table lookup Power demand Logic gates Hardware

来源：评论

学校读者我要写书评

暂无评论

Power efficient FPGA-based TCAM architecture by using segmented matchline strategy

Power efficient FPGA-based TCAM architecture by using segmen...

引用

2019 International Conference on Advances in the Emerging Computing Technologies, AECT 2019

作者： Ur Rehman, Najib Mujahid, Omer Ullah, Zahid Hafeez, Abdul Fouzder, Tama Ibrahim, Muhammad CECOS University of IT Emerging Sciences Department of Electrical Engineering Peshawar Pakistan University of Engineering and Technology Department of Computer Science IT Pakistan University of Liberal Arts Bangladesh Department of Electrical and Electronic Engineering Bangladesh

ISBN: (纸本)9781728144528

Ternary content-addressable memory (TCAM) is famous for its high-speed search operation but this speed comes at the cost of high-power consumption. The paper presents a power efficient architecture for TCAM on field-programmable gate array (FPGA) by using segmented matchline (ML) strategy. Each ML is divided into four equal segments where each segment has nine bits. The proposed TCAM architecture is implemented on Xilinx Virtex-6 FPGA for the size of 64times 36. Implementation results show that during search operation, the proposed TCAM results in 13.88% reduction in dynamic power consumption compared to the latest FPGA-based TCAMs while not compromising on speed. © 2020 IEEE.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

A wearable millimeter Wave MIMO antenna design for high frequency applications

A wearable millimeter Wave MIMO antenna design for high freq...

引用

2019 International Conference on Advances in the Emerging Computing Technologies, AECT 2019

作者： Shafeeq, Muhammad Ibrahim, Muhammad Ullah, Zahid Hafeez, Abdul Fouzder, Tama CECOS University of IT Emerging Sciences Department of Electrical Engineering Peshawar Pakistan University of Engineering and Technology Department of Computer Science IT Pakistan University of Liberal Arts Bangladesh Department of Electrical and Electronic Engineering Bangladesh

ISBN: (纸本)9781728144528

In this paper, we propose an improved methodology for the reduction of mutual coupling within millimeter-wave (mmwave) wearable multiple-input and multiple-output (MIMO) antennas. Such approach is crucial for high frequency applications. The important parameters such as return loss (with best matching having S_{11}, lt -10 dB with 35 GHz centered frequency), radiation pattern, diversity gain (DG gt 9.99), envelop correlation coefficient (ECC lt 0.05), and mutual coupling evaluate the presented antenna. The main issue that arises in MIMO antenna is mutual coupling. We mitigate this problem by introducing slots on either end of the transmission line. We achieve a reduced mutual coupling even less than -36 dB and it works within a high-frequency range, which is its beauty. © 2020 IEEE.

关键词： Millimeter waves

来源：评论

学校读者我要写书评

暂无评论

A Wearable Millimeter Wave MIMO Antenna Design For High Frequency Applications

A Wearable Millimeter Wave MIMO Antenna Design For High Freq...

引用

Advances in the Emerging Computing Technologies (AECT), International Conference on

作者： Muhammad Shafeeq Muhammad Ibrahim Zahid Ullah Abdul Hafeez Tama Fouzder Department of Electrical Engineering CECOS University of IT & Emerging Sciences Peshawar Pakistan Department of Computer Science & IT University of Engineering and Technology Pakistan Department of Electrical and Electronic Engineering University of Liberal Arts Bangladesh Bangladesh

In this paper, we propose an improved methodology for the reduction of mutual coupling within millimeter-wave (mm-wave) wearable multiple-input and multiple-output (MIMO) antennas. Such approach is crucial for high fr... 详细信息

ISBN: (数字)9781728144528

ISBN: (纸本)9781728144535

关键词： antenna radiation patterns microstrip antenna arrays millimetre wave antenna arrays MIMO communication wearable antennas millimetre wave antenna arrays wearable antennas microstrip antenna arrays MIMO communication antenna radiation patterns millimeter wave multiple input multiple output systems High frequency Mutual coupling Wearable

来源：评论

学校读者我要写书评

暂无评论

Forecasting stress, mood, and health from daytime physiology in office workers and students

Forecasting stress, mood, and health from daytime physiology...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Terumi Umematsu Akane Sano Sara Taylor Masanori Tsujikawa Rosalind W. Picard Biometrics Research Laboratories NEC Corporation Japan Department of Electrical and Computer Engineering Rice University Media Lab. Massachusetts Institute of Technology

ISBN: (数字)9781728119908

ISBN: (纸本)9781728119915

We examine the problem of forecasting tomorrow morning's three self-reported levels (on scales from 0 to 100) of stressed-calm, sad-happy, and sick-healthy based on physiological data (skin conductance, skin temperature, and acceleration) from a sensor worn on the wrist from 10am-5pm today. We train automated forecasting regression algorithms using Random Forests and compare their performance over two sets of data: "workers" consisting of 490 days of weekday data from 39 employees at a high-tech company in Japan and "students" consisting of 3,841 days of weekday data from 201 New England USA college students. Mean absolute errors on held-out test data achieved 10.8, 13.5, and 14.4 for the estimated levels of mood, stress, and health respectively of office workers, and 17.8, 20.3, and 20.4 for the mood, stress, and health respectively of students. Overall the two groups reported comparable stress and mood scores, while employees reported slightly poorer health, and engaged in significantly lower levels of physical activity as measured by accelerometers. We further examine differences in population features and how systems trained on each population performed when tested on the other.

关键词： Stress Mood Physiology Feature extraction Forecasting Temperature distribution Skin

来源：评论

学校读者我要写书评

暂无评论

A Novel Defocus-blur Region Detection Approach Based on DCT Feature and PCNN Structure

arXiv

引用

arXiv 2023年

作者： Basar, Sadia Ali, Mushtaq Waheed, Abdul Ahmad, Muneer Miraz, Mahdi H. Department of CS & IT Hazara University Mansehra21120 Pakistan Higher Education Department Govt. Girls Degree College Palosai KPK Peshawar25121 Pakistan Department of Computer Science Women University Swabi Swabi23430 Pakistan School of Electrical and Computer Engineering Seoul National University Seoul08826 Korea Republic of Department of Human and Digital Interface Woosong University Daejeon Korea Republic of School of Computing and Data Science Xiamen University Malaysia Sepang Malaysia School of Computing Faculty of ArtsScience and Technology Wrexham Glyndwr University Wrexham United Kingdom Faculty of Computing Engineering and Science University of South Wales Pontypridd United Kingdom

The motion or out-of-focus effect in digital images is the main reason for the blurred regions in defocused-blurred images. It may adversely affect various image features such as texture, pixel, and region. Therefore, it is important to detect in-focused objects in defocused-blurred images after the segmentation of blurred and non-blurred regions. The state-of-the-art techniques are prone to noisy pixels, and their local descriptors for developing segmentation metrics are also complex. To address these issues, this research, therefore, proposed a novel and hybrid-focused detection approach based on Discrete Cosine Transform (DCT) coefficients and PC Neural Net (PCNN) structure. The proposed approach partially resolves the limitations of the existing contrast schemes to detect in-focused smooth objects from the out-of-focused smooth regions in the defocus dataset. The visual and quantitative evaluation illustrates that the proposed approach outperformed in terms of accuracy and efficiency to referenced algorithms. The highest Fα-score of the proposed approach on Zhao's dataset is 0.7940 whereas on Shi's dataset is 0.9178. © 2023, CC BY.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

On the forward and backward motion of milli-bristle-bots

arXiv

引用

arXiv 2020年

作者： Kim, D. Hao, Z. Mohazab, A.R. Ansari, A. School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States The Foundation for the Advancement of Sciences Humanities Engineering and Arts VancouverBC Canada

This works presents the theoretical analysis and experimental observations of bidirectional motion of a millimeter-scale bristle robot (milli-bristle-bot) with an on-board piezoelectric actuator. First, the theory of the motion, based on the dry-friction model, is developed and the frequency regions of the forward and backward motion, along with resonant frequencies of the system are predicted. Secondly, milli-bristle-bots with two different bristle tilt angles are fabricated, and their bidirectional motions are experimentally investigated. The dependency of the robot speed on the actuation frequency is studied, which reveals two distinct frequency regions for the forward and backward motions that well matches our theoretical predictions. Furthermore, the dependencies of the resonance frequency and robot speed on the bristle tilt angle are experimentally studied and tied to the theoretical model. This work marks the first demonstration of bidirectional motion at millimeter-scales, achieved for bristle-bots with a single on-board actuator. Copyright © 2020, The Authors. All rights reserved.

关键词： Piezoelectricity

来源：评论

学校读者我要写书评

暂无评论

Maneuver at Micro Scale: Steering by Actuation Frequency Control in Micro Bristle Robots*

Maneuver at Micro Scale: Steering by Actuation Frequency Con...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Zhijian Hao DeaGyu Kim Ali Reza Mohazab Azadeh Ansari School of Electrical and Computer Engineering at the Georgia Institute of Technology Atlanta GA USA Foundation for the Advancement of Sciences Humanities Engineering and Arts FASHEA Richmond BC Canada

The following topics are dealt with: mobile robots; learning (artificial intelligence); robot vision; path planning; motion control; manipulators; optimisation; neural nets; collision avoidance; cameras.

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

关键词： Robots Resonant frequency Actuators Vibrations Steady-state Wires Mathematical model

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：