检索结果-内蒙古大学图书馆

Real-Time Systems Symposium (RTSS)

作者： Zonghong Li Guoqi Xie Wenhong Ma Xiongren Xiao Yong Xie Wei Ren Wanli Chang Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University China Research Institute of Hunan University in Chongqing China Nanjing University of Posts and Telecommunications China Huawei Technologies China

Virtual machines (VMs) have been increasingly deployed in real-time systems to enhance heterogeneous resource sharing. Despite being isolated for security and prevention of failure propagation, VMs need to communicate with each other to complete certain tasks jointly. A real-time inter-VM communication framework has been proposed based on VirtIO, which is in essence a mechanism of message passing with high latency and low scalability on data amount. In contrary, IVSHMEM implements inter-VM communication with shared memory, which is generally fast and large in throughput. Unfortunately, IVSHMEM does not resolve resource contention between flows and hence cannot be applied in real-time scenarios. In this work, we propose a real-time inter-VM communication framework RTISM built upon shared memory and IVSHMEM. RTISM provides priority-based scheduling and supports mixed-criticality flows. Worst-Case Response Time (WCRT) analysis is reported to bound the end-to-end communication delay and a Limited Priority Assignment (LPA) algorithm is developed to enhance schedulability. Experimental evaluation shows that (i) RTISM has about 4 times higher throughput than the VirtIO-based inter-VM communication framework; (ii) LPA improves schedulability by over 25% compared to the state-of-the-art; (iii) WCRT produces a tight bound.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Stacking-Based Attention Temporal Convolutional network for Action Segmentation 48

Stacking-Based Attention Temporal Convolutional Network for ...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Yang, Liu Jiang, Yu Hong, Junkun Wu, Zhenjie Yang, Zhan Long, Jun Central South University School of Computer Science and Engineering China Central South University Big Data Institute Central South University China Central South University Network Resources Management and Trust Evaluation Key Laboratory of Hunan Province Changsha China

ISBN: (纸本)9781728163277

Action segmentation plays an important role in video understanding, which is implemented by frame-wise action classification. Recent works on action segmentation capture long-term dependencies by increasing temporal convolution layers in Temporal Convolution networks (TCNs). However, high layers in TCNs are more coarse access to video features, resulting in the loss of fine-grained information for frame-wise action classification. To address the above issues, we propose a novel Attention-based Temporal Convolution (ATC) block to capture fine-grained information of temporal dependencies for frame-wise action classification by self-attention mechanism. Via stacking ATC blocks, we design a Stacking-based Attention Temporal Convolutional network (SATC) to adaptively capture long-term and short-term dependencies, according to the semantic similarity of features on different temporal receptive fields simultaneously. The experimental results demonstrate that our SATC outperforms other baselines on all three challenging datasets: GTEA, 50Salads and Breakfast. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Study on data de-duplication based on bloom filter in fault tolerant erasure codes

引用

Journal of Computational and Theoretical Nanoscience 2016年第1期13卷 179-185页

作者： Yan, Xiai Zhang, Dafang Zhang, Boyun College of Computer Science and Elctronic Engineering Hunan University 410082 China Department of Information Technology Hunan Police Academy Changsha410138 China Key Laboratory of Network Crime Investigation Colleges of Hunan Province Changsha410138 China

Erasure codes have become a mainstream fault-tolerant method in the field of cloud storage fault tolerance. Erasure codes will be the data from the file level storage into a block level storage, there will be a lot of repeat encoding block. Data De-duplication technique provide an effective solution to the minimization of storage. Bloom filter is a kind of space efficient data structure, which is suitable for the query in finite space. To solve the problem of erasure codes repetition coding block, it proposes a global bloom filter array efficient methods, and through the dynamic expansion of the standard Bloom filters and set up a global dynamic bloom filter array GDBFA, so as to effectively control the system false positive rate. Experiments show that: GDBFA has the characteristics of high De-duplication rate, low false positive rate, and less influence on the performance of fault tolerance. © 2016 American Scientific Publishers All rights reserved.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

A delay-derivative-dependent approach to robust H ∞ filter design for uncertain systems with time-varying distributed delays

引用

Journal of the Franklin Institute 2011年第2期348卷 179-200页

作者： Jiyao An Guilin Wen Nianfei Gan Renfa Li State Key Laboratory of Advanced Design and Manufacture for Vehicle Body the Key Laboratory of Advanced Design and Simulation Techniques for Special Equipment Ministry of Education Hunan University Changsha 410082 PR China School of Computer and Communication Institute of Embedded Systems and Network Hunan University Changsha 410082 PR China

This paper focuses on the problem of robust H ∞ filter design for uncertain systems with time-varying state and distributed delays. System uncertainties are considered as norm-bounded time-varying parametric uncertainties . The delays are assumed to be time-varying delays being differentiable uniformly bounded with delay-derivative bounded by a constant, which may be greater than one. A new delay-derivative-dependent approach of filter design for the systems is proposed. A novel Lyapunov–Krasovskii functional (LKF) is employed, and a tighter upper bound of its derivative is obtained by employing an inequality and using free-weighting matrices technique, then the proposed result has advantages over some existing results, in that it has less conservatism and it enlarges the application scope. An improved sufficient condition for the existence of such a filter is established in terms of linear matrix inequality (LMI). Finally, illustrative examples are given to show the effectiveness and reduced conservatism of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Real-Time CAN-CAN Gateway with Tight Latency Analysis and Targeted Priority Assignment

A Real-Time CAN-CAN Gateway with Tight Latency Analysis and ...

引用

Real-Time Systems Symposium (RTSS)

作者： Guoqi Xie Haijie Gong Yunbo Han Samarjit Chakraborty Wanli Chang College of Computer Science and Electronic Engineering Hunan University China Key Laboratory for Embedded and Network Computing of Hunan Province China Future Network Lab CSIG Tencent China University of North Carolina at Chapel Hill USA University of York UK

ISBN: (数字)9781728183244

ISBN: (纸本)9781728183251

There is a demand in the automotive industry to connect two CAN-based subsystems. The commercial CAN-CAN gateway supports basic message forwarding with no real-time behavior. To address this issue, a new gateway architecture is described, on which we present a novel worst-case latency analysis. Specifically, we bound the arrival of the messages at the gateway, which is then used by the Pointer Reachability Exploration (PRE) to derive the interfering message jobs. Our analysis computes a safe gateway latency tighter than the conventional one applied in CAN. Furthermore, we propose a Targeted Priority Assignment (TPA) algorithm that targets at the priorities assigned at the CAN bus and runs a reordering at the gateway to enhance the schedulability. TPA performs better than DMPO (Deadline Monotonic Priority Ordering), while OPA (Audsley's Optimal Priority Assignment) cannot be applied in this context. Evaluation over real-life and scalable CAN message sets is conducted. The reported analysis and priority assignment algorithm are developed for dynamic use to improve the acceptance ratio and can also be deployed statically to provide timing guarantees. This work can be easily extended to support multiple CAN subsystems.

关键词： Heuristic algorithms Prototypes Computer architecture Logic gates Real-time systems Timing Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Privacy preserving for network coding in smart grid 15th

Privacy preserving for network coding in smart grid

引用

15th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2015

作者： He, Shiming Zeng, Weini Xie, Kun Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation School of Computer and Communication Engineering Hunan Province Engineering Research Center of Electric Transportation and Smart Distributed Network Changsha University of Science and Technology Changsha410114 China The 716th Research Institute China Shipbuilding Industry Corporation Lianyungang222061 China College of Computer Science and Electronics Engineering Hunan University Changsha410082 China Department of Electrical and Computer Engineering State University of New York at Stony Brook New York United States

ISBN: (纸本)9783319271361

In smart grid, privacy implications to individuals and their family is an important issue, due to the fine-grained usage data collection. Wireless communications are considered by many utility companies to obtain information. network coding is exploited in smart grids, to enhance network performance in terms of throughput, delay, robustness, and energy consumption. However, network Coding introduces new challenge for privacy preserving due to the encoding of packet in forwarder nodes. We propose a distributed privacy preserving scheme for network coding in smart grid, which considers the converged flows character of smart grid and exploits a homomorphic encryption function to decrease the complex in forwarder node. The message content of packet is encrypted and the tag of packet is encrypted by homomorphic encryption function. Then the forwarder node linear random codes the encrypted message contents and directly processes the tags cryptotext based on the homomorphism feature. It offers message content confidentiality privacy preserving feature, which can efficiently thwart traffic analysis. Extensive security analysis and performance evaluations demonstrate the validity and efficiency of the proposed scheme. © Springer International Publishing Switzerland 2015.

关键词： Smart power grids

来源：评论

学校读者我要写书评

暂无评论

Efficient group signature scheme without pairings

引用

International Journal of network Security 2020年第3期22卷 504-515页

作者： Gu, Ke Liu, Dianxing Yin, Bo School of Computer and Communication Engineering Changsha University of Science and Technology Wangjiali Rd Tianxin district Changsha Hunan Province410114 China Key Laboratory of Network Crime Investigation of Hunan Provincial Colleges Hunan Police Academy Wangjiali Rd Tianxin district Changsha Hunan Province410114 China

Although currently many group signature schemes have been proposed, most of them are constructed on pair-ings. In this paper, we present an efficient group signature scheme without pairings under the model of verifier-local revocation, which is based on the modified EDL signature (first proposed by D. Chaum et al. in Crypto 92). Com-pared with other group signature schemes, the proposed scheme does not employ pairing computation and has the constant signing time and signature size, whose secu-rity can be reduced to the computational Diffie-Hellman (CDH) assumption in the random oracle model. Also, we give a formal security model for group signature and prove that the proposed scheme has the properties of traceabil-ity and anonymity. © 2020 Femto Technique Co., Ltd.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

TS-ENAS:Two-Stage Evolution for Cell-based network Architecture Search

arXiv

引用

arXiv 2023年

作者： Zou, Juan Wu, Shenghong Xia, Yizhang Jiang, Weiwei Wu, Zeping Zheng, Jinhua Engineering Research Center of Hunan Province for Optimization and Security of Intelligent System Key Laboratory of Intelligent Computing and Information Processing Ministry of Education of China Key Laboratory of Hunan Province for Internet of Things and Information Security Xiangtan University Hunan Province Xiangtan411105 China The College of Aerospace Science and Engineering National University of Defense Technology Hunan Changsha410073 China

Neural network architecture search provides a solution to the automatic design of network structures. However, it is difficult to search the whole network architecture directly. Although using stacked cells to search neural network architectures is an effective way to reduce the complexity of searching, these methods do not able find the global optimal neural network structure since the number of layers, cells and connection methods is fixed. In this paper, we propose a Two-Stage Evolution for cell-based network Architecture Search(TS-ENAS), including one-stage searching based on stacked cells and second-stage adjusting these cells. In our algorithm, a new cell-based search space and an effective two-stage encoding method are designed to represent cells and neural network structures. In addition, a cell-based weight inheritance strategy is designed to initialize the weight of the network, which significantly reduces the running time of the algorithm. The proposed methods are extensively tested and compared on four image classification dataset, Fashion-MNIST, CIFAR10, CIFAR100 and ImageNet and compared with 22 state-of-the-art algorithms including hand-designed networks and NAS networks. The experimental results show that TS-ENAS can more effectively find the neural network architecture with comparative performance. Copyright © 2023, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

A deep hashing method based on attention module for image retrieval

A deep hashing method based on attention module for image re...

引用

International Conference on Intelligent Computation Technology and Automation, ICICTA

作者： Jun Long Xiangxiang Wei Qianqian Qi Ye Wang The School of Computer Science and Engineering Central South University Changsha China Network Resources Management and Trust Evaluation Key Laboratory of Hunan Province Changsha China

Hashing retrieval technique has received widespread attention since its low storage consumption and fast retrieval speed. Image retrieval based on deep hashing method uses image features with high-level semantic information and discriminative expression capabilities. However, there are still some limitations: (1) Compared with the traditional feature vector, the length of the learned hash codes is short, and each bit of the code is -1 or 1, the discrimination capability of feature representation is limited; (2) the existing deep hash algorithms cannot directly learn the discrete hash codes using sign activation function, therefore, they use relaxation-based scheme to learn the hash codes, resulting in a large quantization error. In order to solve the above problems, a deep hashing retrieval algorithm incorporate with attention model is proposed. This algorithm embeds spatial and channel attention models in the feature extraction network, and uses a novel activation function in the hash layer. In addition, a novel triple loss function is further proposed to improve the learning capability of the network. Finally, experimental results on the two benchmark databases, i.e., CIFAR-10 and MINIST datasets, show the effectiveness and superiority compared with the baselines.

关键词： Codes Quantization (signal) Automation Databases Computational modeling Image retrieval Semantics

来源：评论

学校读者我要写书评

暂无评论

Scholars influence evaluation based on time series heterogeneous network

Scholars influence evaluation based on time series heterogen...

引用

International Conference on Measuring Technology and Mechatronics Automation, ICMTMA

作者： Ye Wang Zhen Ding Xiang Xiang Wei Jun Long The School of Computer Science and Engineering Central South University Changsha China Network Resources Management and Trust Evaluation Key Laboratory of Hunan Province Changsha China

In recent years, with the rapid development of network and data mining, multi-source heterogeneous data related to academics has also shown an exponential growth. As an important research field of academic big data, the objectivity of academic influence evaluation methods and rationality plays an important role in the evaluation of scientific and technological talents, academic groups, and national scientific research strength. Traditional methods are generally measured based on the amount of papers published by scholars and the total frequency of citations. Although these methods are simple to calculate the influence of scholar, they ignore the importance of the cooperative network of scholar and the evolution of academic networks over time. In order to solve the above limitations, this paper constructs a derivation of scholars' paper relationship network and scholars' cooperation network, taking into account the importance of time in the network, from the perspectives of the journal where the scholar publishes papers, the location of the author, and the number of citations of the paper. In this paper, a method is designed to calculate the academic influence of the scholars, making the calculation result more reasonable.

关键词： Computer science Mechatronics Automation network topology Time series analysis Cooperative systems Big Data

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：