检索结果-内蒙古大学图书馆

37th IEEE International conference on computer Design (ICCD)

作者： Chen, Jianqi Schafer, Benjamin Carrion Univ Texas Dallas Dept Elect & Comp Engn Richardson TX 75083 USA

ISBN: (纸本)9781538666487

This paper presents a method to generate a variety of micro-architectures for a given hardware accelerator mapped onto reconfigurable fabric optimized for different operating frequencies. The most optimal micro-architecture is then loaded onto the fabric for a given operating frequency in order to minimize the power consumption. State-of-the-art FPGAs are runtime reconfigurable and provide multiple clock domains. This enables these devices to reconfigure any accelerator mapped on them and their frequencies at runtime. At the same time, FPGA vendors have embraced high-Level Synthesis (HLS) to increase the design productivity and help designers with limited hardware development skills to program these devices. One of the advantages of HLS over traditional RT-level hardware design is that HLS allows to automatically generate micro-architectures with unique area, performance and power trade-offs by setting different synthesis options, which is impractical or very time consuming at the RT-level. This work leverages these two features and investigates the benefit of adapting the micro-architecture of hardware accelerators mapped onto a reconfigurable fabric at runtime when the operating frequency changes to reduce the power consumption, while maximizing the throughput. To enable the frequency-aware micro-architectural adaptation we also propose a simple micro-architectural resource manager and show that the overhead in terms of area and delay is negligible. We conduct two sets of experiments. The first shows that our proposed approach leads to faster circuits which consume less power than just statically scaling the frequency of the fastest micro-architecture for a variety of different test cases. The second case, presents case study of a face detection application mapped onto a battery-operated wireless camera sensor node powered by solar cells.

关键词： Dynamic Frequency Scaling high level Synthesis Low Power FPGA design Partial Reconfiguration Runtime Micro architectural Adaptation

来源：评论

学校读者我要写书评

暂无评论

high-performance and Resource-Efficient IoT-based Sleep Monitoring System

High-Performance and Resource-Efficient IoT-based Sleep Moni...

引用

IEEE conference on Vehicular Technology (VTC)

作者： Nico Surantha Oei Kurniawan Utomo Sani Muhamad Isa Computer Science Department BINUS Graduate Program-Master of Computer Science Jakarta Indonesia

ISBN: (数字)9781728152073

ISBN: (纸本)9781728140537

Recently, the number of people who have trouble sleeping is on the rise. It results in an increase in the number of car accidents, lower productivity and a higher risk of chronic disease. Advancing wearable and contactless health sensors offers the possibility of daily sleep quality monitoring that can be done on a regular basis by patients in their homes. The goal of this research is to propose the design of an Internet-of-Things (IoT) sleep quality monitoring system that can be integrated with a wearable or contactless sensor. The proposed platform architecture is based on micro-services and is event-driven. The simulation results show that the implementation of the IoT sleep quality monitoring system based on the proposed design decreases the response time by 55.85 per cent while increasing the throughput by 34.76 per cent compared to the traditional approach. On the other hand, the rate of increase in memory usage per instance replication for proposed method is lower than the traditional method.

关键词： Monitoring Logic gates Time factors computer architecture Sensor systems Throughput

来源：评论

学校读者我要写书评

暂无评论

Adaptive Interference-Aware VNF Placement for Service-Customized 5G Network Slices

Adaptive Interference-Aware VNF Placement for Service-Custom...

引用

IEEE conference on computer Communications (IEEE INFOCOM)

作者： Zhang, Qixia Liu, Fangming Zeng, Chaobing Huazhong Univ Sci & Technol Sch Comp Sci & Technol Key Lab Serv Comp Technol & Syst Minist Educ Wuhan Hubei Peoples R China

ISBN: (纸本)9781728105154

Based on network function virtualization (NFV) and software defined network (SDN), network slicing is proposed as a new paradigm for building service-customized 5G network. In each network slice, service-required virtual network functions (VNFs) can be flexibly deployed in an on-demand manner, which will support a variety of 5G use cases. However, due to the diverse performance requirements among different 5G scenarios, an adaptive VNF placement approach is needed to automatically accommodate to service-specific requirements. In this paper, we tackle the VNF placement problem by first proposing a general 5G network slice framework, which jointly contains both edge cloud and core cloud servers. Specially, based on the fact that VNF consolidation may cause severe performance degradation, we adopt a demand-supply model to quantify the VNF interference. With an aim to maximize the total throughput of accepted requests, we propose an Adaptive Interference-Aware (AIA) heuristic approach to automatically place VNFs in 5G service-customized network slices. Through simulations on two typical 5G scenarios, we demonstrate that AIA can efficiently handle traffic variation especially caused by VNF interference and improve the total throughput by 20.11% and 24.21% in autonomous driving and 4K/SK HD video network slices as compared with the state-of-the-art methods.

关键词： 5G mobile communication Servers Interference Throughput Autonomous vehicles high definition video Adaptive systems

来源：评论

学校读者我要写书评

暂无评论

SIMPLE RECUPERATED S-CO2 CYCLE REVISITED: OPTIMIZATION OF OPERATING PARAMETERS FOR MAXIMUM CYCLE EFFICIENCY

SIMPLE RECUPERATED S-CO2 CYCLE REVISITED: OPTIMIZATION OF OP...

引用

ASME Turbo Expo: Turbomachinery Technical conference and Exposition

作者： Dutta, Anchit Gupta, Adhip Sathish, Sharath Bandooni, Aman Kumar, Pramod RV Coll Engn Dept Mech Engn Bangalore Karnataka India Indian Inst Sci Dept Mech Engn Bangalore Karnataka India

ISBN: (纸本)9780791858721

The paper presents modeling and Design of Experiments (DOE) analysis for a simple recuperated s-CO2 closed loop Brayton cycle operating at a maximum temperature of 600 degrees C and a compressor inlet temperature of 45 degrees C. The analysis highlights the impact of isentropic efficiencies of the turbine and compressor, decoupled in this case, on other equipment such as recuperator, gas cooler and heater, all of which have a bearing on the overall performance of the s-CO2 Brayton cycle. A MATLAB program coupled with REFPROP is used to perform the thermodynamic analysis of the cycle. A design space exploration with a Design of Experiments (DOE) study is undertaken using I-sight (TM) (multi-objective optimization software), which is coupled with the MATLAB code. The outcome of the DOE study provides the optimal pressure ratios and high side pressures for maximum cycle efficiency in the design space. By varying pressure ratios along with a floating high side pressure, the analysis reveals that the cycle performance exhibits a peak around a pressure ratio of 2.5, with cycle efficiency being the objective function. A further interesting outcome of the DOE study reveals that the isentropic efficiencies of the compressor and turbine have a strong influence not only on the overall cycle efficiency, but also the optimum pressure ratio as well as the threshold pressures (low as well as high side pressure). An important outcome of this exercise shows that the isentropic efficiency of the turbine has a much greater impact on the overall cycle performance as compared to that of the compressor.

关键词： Optimization Cycles Supercritical carbon dioxide Pressure Compressors performance Turbines Brayton cycle Temperature Design Experimental design Matlab Pareto optimization Modeling Bearings computer software

来源：评论

学校读者我要写书评

暂无评论

A Fourier Perspective on Model Robustness in computer Vision 33

A Fourier Perspective on Model Robustness in Computer Vision

引用

33rd conference on Neural Information Processing Systems (NeurIPS)

作者： Yin, Dong Lopes, Raphael Gontijo Shlens, Jonathon Cubuk, Ekin D. Gilmer, Justin Univ Calif Berkeley Dept EECS Berkeley CA 94720 USA Google Res Brain Team Mountain View CA 94043 USA Google AI Residency Program Redwood City CA USA

Achieving robustness to distributional shift is a longstanding and challenging goal of computer vision. Data augmentation is a commonly used approach for improving robustness, however robustness gains are typically not uniform across corruption types. Indeed increasing performance in the presence of random noise is often met with reduced performance on other corruptions such as contrast change. Understanding when and why these sorts of trade-offs occur is a crucial step towards mitigating them. Towards this end, we investigate recently observed trade-offs caused by Gaussian data augmentation and adversarial training. We find that both methods improve robustness to corruptions that are concentrated in the high frequency domain while reducing robustness to corruptions that are concentrated in the low frequency domain. This suggests that one way to mitigate these trade-offs via data augmentation is to use a more diverse set of augmentations. Towards this end we observe that AutoAugment [6], a recently proposed data augmentation policy optimized for clean accuracy, achieves state-of-the-art robustness on the CIFAR-10-C [17] benchmark.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Implementation with comparison of system performance in different parallel processing configuration systems using matrix multiplication

引用

AIP conference Proceedings 2022年第1期2386卷

作者： Dhurgham A. Habeeban Yahya M. Al-Mayali 1University of Kufa Faculty of Computer Science and Math Department of Computer Science Al-Najaf Al-Ashraf Iraq 2University of Alkafeel College of Engineering Techniques Al-Najaf Al-Ashraf Iraq

Nowadays the Parallel Processing Architecture became a public phenomenon in computer community and Information Technology, in both fields: the Hardware, and the Software. Where most applications, which are using IT, make use and gain beneficial usage from the facilities that are provided by this advance computing approach. This paper is an attempt that has been made to establish, analyze, implement, and compare an environment for parallel Processing system that uses the available computer equipment and the software. One of the most main targets of parallel processing is to solve large mathematical problems such as matrix multiplication and sorting large numbers of records. This type of large problems requires reasonable reasonably high processing time. Most modern high level programming languages such as Fortran, C++, MATLAB, Python, C#, Java, .etc., have programming constructs that deal with the parallel processing. In this research paper, the programming language that has been chosen is C# and the experiment is the multiplication of two matrices with varying size and running the program on different configuration systems, to see the effect of increasing the number of nodes and changing method of parallel processing: Serial, parallel, Distributed over LAN. As a result of the experiment of two matrices of the size 8000 x 8000, and executing the parallel program on different configurations of parallel system show that the distributed system using LAN achieved better result where the execution is equal to 935.183 seconds, while with multithreaded single parallel machine=8772.732 seconds and using serial machine=14755.71 seconds.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Accurate Semidefinite Relaxation Method for 3-D Rigid Body Localization Using AOA

Accurate Semidefinite Relaxation Method for 3-D Rigid Body L...

引用

IEEE International conference on Acoustics, Speech and Signal Processing

作者： Gang Wang K. C. Ho Faculty of Electrical Engineering and Computer Science Ningbo University China Electrical Engineering and Computer Science Department University of Missouri USA

ISBN: (数字)9781509066315

ISBN: (纸本)9781509066322

This paper addresses the rigid body localization problem using angle-of-arrival measurements. We formulate the problem as a constrained weighted least squares (CWLS) minimization problem with the rotation matrix and position vector as variables, which is a challenging non-convex problem. To approximately solve this problem, we first relax it as a convex semidefinite program (SDP), and then tighten the relaxed problem by adding some reasonable second-order cone constraints. Simulations show that the tightened SDP problem is able to reach the performance of the original CWLS problem, making its solution achieve the Cramer-Rao lower bound accuracy, when the noise level is not too high.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DPA: Demand-Based Partition and Data Allocation for Hybrid On-Chip Memory

DPA: Demand-Based Partition and Data Allocation for Hybrid O...

引用

IEEE International conference on computer and Communication Engineering Technology (CCET)

作者： Jianfeng Wang Leibo Liu Jianfeng Zhu Shaojun Wei Institute of Microelectronics Tsinghua University Beijing China

ISBN: (数字)9781728188119

ISBN: (纸本)9781728188126

The performance and energy consumption of the memory system has become an essential bottleneck in modern processors. Scratchpad memory (SPM) is widely used in the embedded system for meeting the constrict of energy consumption. The hybrid memory architecture has been proposed to improve the energy efficiency of the overall system. Previous works focus on optimizing data allocations in the fixed-size cache and SPM. However, applications have different preferences for cache and SPM. Fixed memory structures may cause a waste of storage resources. In this paper, a reconfigurable mechanism is proposed to dynamically adjust the partition of the local memory and allocate data with high-locality into the SPM. Our experimental results show that our method improves the energy-delay product by 12.0% on average compared with the previous works.

关键词： Resource management Memory management Runtime History Energy consumption program processors

来源：评论

学校读者我要写书评

暂无评论

Rhino: An Open-source Embedded Motherboard Design Enabling Complex Behavior of Intelligent Robots

Rhino: An Open-source Embedded Motherboard Design Enabling C...

引用

IEEE/ASME International conference on Advanced Intelligent Mechatronics (AIM)

作者： Gao, Yucen Gong, Liang Huang, Yixiang Liu, Chengliang Li, Xinyang Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Comp Sci & Technol Shanghai Peoples R China Shanghai Jiao Tong Univ Shanghai Peoples R China Shanghai Jia Ye Elect Technol Co Ltd Shanghai Peoples R China

ISBN: (纸本)9781728124933

In recent years, Robot Operating System (ROS) has become a de facto standard for many robotic systems. However, there lacks a general-purpose control hardware to perfectly support ROS applications in an embedded fashion. In this paper, we take a hardware/software co-design methodology and a loosely coupled design methodology to develop a ROS-oriented motherboard dedicatedly for facilitating high-end intelligent robotic applications. First, orienting around the ROS computational graph level, the hardware/software co-design is proposed to realize a mirrored modular design paradigm. Second, an open-source ROS motherboard, namely the "Rhino", is accordingly designed with the highlight of accelerating the embedded neuromorphic computation. Third, real-time performance and feasibility of Rhino are validated at different scales. Experimentation shows that the open-source prototype motherboard is eligible for ROS-based robot development and outperforms the conventional IPC and tailor-made control board. ROS-oriented hardware/software co-design paradigm complements the ROS ecosystem with an open-source AI-enabled motherboard for developing intelligent robots.

关键词： Hardware Robots Open source software computer architecture Operating systems Real-time systems

来源：评论

学校读者我要写书评

暂无评论

SRGAN with Total Variation Loss in Face Super-Resolution

SRGAN with Total Variation Loss in Face Super-Resolution

引用

National Foundation for Science and Technology Development conference on Information and computer Science (NICS)

作者： Hai Nguyen-Truong Khoa N. A. Nguyen San Cao Advanced Program in Computer Science Faculty of Information Technology Vietnam National University Ho Chi Minh city Vietnam

ISBN: (数字)9780738105536

ISBN: (纸本)9781665405188

Facial image super-resolution is a crucial preprocessing for facial image analysis, face recognition, and image-based 3D face reconstruction. Convolutional neural networks were earlier used to produce high-resolution images that train quicker and shown excellent performance by learning mapping relation using pairs of low-resolution and high-resolution images. However, in some cases, they are incapable of recovering finer details and often generate blurry images. In this paper, we evaluate a method of applying Generative adversarial networks in generating realistic super-resolution images from low-resolution ones by using three typical losses for super-resolution: Content Loss, Adversarial Loss, Perceptual Loss, and proposed to use Total Variation Loss. We try different pre-trained famous Convolutional neural networks models (VGG19, FaceNet, and EfficientNet) in Perceptual Loss to have a general view with different backbones. Our network gains 32.67 of Peak signal-to-noise ratio (PSNR) and 0.89 of Structural similarity index (SSIM) in 100 random samples from the Flickr-Faces-HQ dataset.

关键词： Training Visualization PSNR Face recognition Superresolution Feature extraction Convolutional neural networks

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：