检索结果-内蒙古大学图书馆

Lightweight Task-Oriented Semantic Communication Empowered by Large-Scale AI Models

IEEE Transactions on Vehicular Technology 2025年

作者： Liu, Chuanhong Guo, Caili Yang, Yang Chen, Mingzhe Quek, Tony Q. S. Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing100876 China Beijing University of Posts and Telecommunications Beijing Laboratory of Advanced Information Networks School of Information and Communication Engineering Beijing100876 China University of Miami Department of Electrical and Computer Engineering Institute for Data Science and Computing Coral GablesFL United States Singapore University of Technology and Design Dept. of Information Systems Technology and Design 487372 Singapore

Recent studies have focused on leveraging large-scale artificial intelligence (LAI) models to improve semantic representation and compression capabilities. However, the substantial computational demands of LAI models pose significant challenges for real-time communication scenarios. To address this, this paper proposes utilizing knowledge distillation (KD) techniques to extract and condense knowledge from LAI models, effectively reducing model complexity and computation latency. Nevertheless, the inherent complexity of LAI models leads to prolonged inference times during distillation, while their lack of channel awareness compromises the distillation performance. These limitations make standard KD methods unsuitable for task-oriented semantic communication scenarios. To address these issues, we propose a fast distillation method featuring a pre-stored compression mechanism that eliminates the need for repetitive inference, significantly improving efficiency. Furthermore, a channel adaptive module is incorporated to dynamically adjust the transmitted semantic information based on varying channel conditions, enhancing communication reliability and adaptability. In addition, an information bottleneck-based loss function is derived to guide the fast distillation process. Simulation results verify that the proposed scheme outperform baselines in term of task accuracy, model size, computation latency, and training data requirements. © 1967-2012 IEEE.

关键词： NP-hard

来源：评论

学校读者我要写书评

暂无评论

Warp-level divergence in GPUs: Characterization, impact, and mitigation

Warp-level divergence in GPUs: Characterization, impact, and...

引用

20th IEEE International Symposium on High Performance Computer architecture, HPCA 2014

作者： Xiang, Ping Yang, Yi Zhou, Huiyang Dept. of Electrical and Computer Engineering North Carolina State University Raleigh NC United States Dept. of Computing Systems Architecture NEC Laboratories America Princeton NJ United States

ISBN: (纸本)9781479930975

High throughput architectures rely on high thread-level parallelism (TLP) to hide execution latencies. In state-of-art graphics processing units (GPUs), threads are organized in a grid of thread blocks (TBs) and each TB contains tens to hundreds of threads. With a TB-level resource management scheme, all the resource required by a TB is allocated/released when it is dispatched to / finished in a streaming multiprocessor (SM). In this paper, we highlight that such TB-level resource management can severely affect the TLP that may be achieved in the hardware. First, different warps in a TB may finish at different times, which we refer to as 'warp-level divergence'. Due to TB-level resource management, the resources allocated to early finished warps are essentially wasted as they need to wait for the longest running warp in the same TB to finish. Second, TB-level management can lead to resource fragmentation. For example, the maximum number of threads to run on an SM in an NVIDIA GTX 480 GPU is 1536. For an application with a TB containing 1024 threads, only 1 TB can run on the SM even though it has sufficient resource for a few hundreds more threads. To overcome these inefficiencies, we propose to allocate and release resources at the warp level. Warps are dispatched to an SM as long as it has sufficient resource for a warp rather than a TB. Furthermore, whenever a warp is completed, its resource is released and can accommodate a new warp. This way, we effectively increase the number of active warps without actually increasing the size of critical resources. We present our lightweight architectural support for our proposed warp-level resource management. The experimental results show that our approach achieves up to 76.0% and an average of 16.0% performance gains and up to 21.7% and an average of 6.7% energy savings at minor hardware overhead. © 2014 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Obtaining ship trajectories of an autonomous in-scale fast-ferry by identifying a heading model

Obtaining ship trajectories of an autonomous in-scale fast-f...

引用

IEEE International Symposium on Intelligent Signal Processing

作者： Velasco, F. J. Revestido, E. Lopez, E. Moyano, E. Dept. Electronics Technology and Systems Engineering and Automatic Control Univ. Cantabria Dept. Sciences and Navigation Technics Engines and Naval Architecture Univ. País Vasco (UPV/EHU) Dept. Applied Mathematics and Computing Sciences Univ. Cantabria

ISBN: (纸本)9781424408290

The aim of this paper is to obtain a model for an in-scale fast-ferry model TF-120 including the dynamics and kinematics, using a remote experimentation platform for marine vehicles. The physical model is autonomous and is controlled remotely from a PC using Wi-Fi communications. The identification and validation of the dynamical model is obtained with turning circle maneuverings. In this model are included the kinematical equations. The tests with the autonomous in-scale physical model of the high-speed ship were carried out in the surroundings of the Bay of Santander. With this model. simulations with several turbojets angles are made.

关键词： parametric identification ship physical model manoeuvring heading idemification

来源：评论

学校读者我要写书评

暂无评论

Introducing SLAMBench, a performance and accuracy benchmarking methodology for SLAM

Introducing SLAMBench, a performance and accuracy benchmarki...

引用

2015 IEEE International Conference on Robotics and Automation, ICRA 2015

作者： Nardi, Luigi Bodin, Bruno Zia, M. Zeeshan Mawer, John Nisbet, Andy Kelly, Paul H. J. Davison, Andrew J. Luján, Mikel O'Boyle, Michael F. P. Riley, Graham Topham, Nigel Furber, Steve Dept. Computing Imperial College London United Kingdom Institute for Computing Systems Architecture Univ. of Edinburgh United Kingdom School of Computer Science Univ. of Manchester United Kingdom

ISBN: (纸本)9781479969234

Real-time dense computer vision and SLAM offer great potential for a new level of scene modelling, tracking and real environmental interaction for many types of robot, but their high computational requirements mean that use on mass market embedded platforms is challenging. Meanwhile, trends in low-cost, low-power processing are towards massive parallelism and heterogeneity, making it difficult for robotics and vision researchers to implement their algorithms in a performance-portable way. In this paper we introduce SLAMBench, a publicly-available software framework which represents a starting point for quantitative, comparable and validatable experimental research to investigate trade-offs in performance, accuracy and energy consumption of a dense RGB-D SLAM system. SLAMBench provides a KinectFusion implementation in C++, OpenMP, OpenCL and CUDA, and harnesses the ICL-NUIM dataset of synthetic RGB-D sequences with trajectory and scene ground truth for reliable accuracy comparison of different implementation and algorithms. We present an analysis and breakdown of the constituent algorithmic elements of KinectFusion, and experimentally investigate their execution time on a variety of multicore and GPU-accelerated platforms. For a popular embedded platform, we also present an analysis of energy efficiency for different configuration alternatives. © 2015 IEEE.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Trustworthy selection of cloud providers based on security and privacy requirements: Justifying trust assumptions

Trustworthy selection of cloud providers based on security a...

引用

10th International Conference on Trust, Privacy and Security in Digital Business, TrustBus 2013

作者： Pavlidis, Michalis Mouratidis, Haralambos Kalloniatis, Christos Islam, Shareeful Gritzalis, Stefanos School of Architecture Computing and Engineering University of East London United Kingdom Cultural Informatics Laboratory Dept. of Cultural Technology and Communication University of the Aegean Greece Laboratory of Information and Communication Systems Security Dept. of Information and Communications Systems Engineering University of the Aegean Greece

ISBN: (纸本)9783642403422

Cloud computing is a new paradigm with a promising potential. However, issues of security, privacy, and trust raise concerns and discourage its adoption. In previous work we presented a framework for the selection of appropriate cloud provider based on security and privacy requirements criteria. However, the adoption of cloud includes release of control over valuable assets, which constitutes trust in the cloud provider of paramount importance. In this paper we extend the framework by incorporating trust and control concepts in its language and adding a new activity to properly identify and reason about trust assumptions during the selection of appropriate cloud provider. Also, the CASE tool was extended to support the new activity. A case study is used to illustrate the usefulness of our approach. © 2013 Springer-Verlag.

关键词： Trusted computing

来源：评论

学校读者我要写书评

暂无评论

Improving monaural speaker identification by double-talk detection

Improving monaural speaker identification by double-talk det...

引用

作者： Saeidi, Rahim Mowlaee, Pejman Kinnunen, Tomi Tan, Zheng-Hua Christensen, Mads Græsbøll Jensen, Søren Holdt Fränti, Pasi School of Computing University of Eastern Finland Joensuu Finland Dept. of Electronic Systems Aalborg University Denmark Architecture Design and Media Technology Aalborg University Denmark

This paper describes a novel approach to improve monoaural speaker identification where two speakers are present in a single-microphone recording. The goal is to identify both of the underlying speakers in the given mixture. The proposed approach is composed of a double-talk detector (DTD) as a preprocessor and speaker identification back-end. We demonstrate that including the double-talk detector improves the speaker identification accuracy. Experiments on GRID corpus show that including the DTD improves average recognition accuracy from 96.53% to 97.43%. © 2010 ISCA.

关键词： Loudspeakers

来源：评论

学校读者我要写书评

暂无评论

SmartSLAM: Localization and mapping across multi-environments

SmartSLAM: Localization and mapping across multi-environment...

引用

2004 IEEE International Conference on systems, Man and Cybernetics, SMC 2004

作者： Asmar, Daniel C. Zelek, John S. Abdallah, Samer M. School of Engineering Eng. Systems and Computing University of Guelph Guelph Ont. Canada Faculty of Eng. and Architecture Dept. of Mechanical Engineering American University of Beirut Beirut Lebanon

ISBN: (纸本)0780385667

In the absence of absolute localization tools such as GPS, a robot can still successfully navigate by conducting Simultaneous Localization and Mapping (SLAM). All SLAM algorithms to date can only be applied in one environment at a time. In this paper we propose to extend SLAM to multi-environments. In SmartSLAM, the robot first classifies its entourage using environment recognition code and then performs SLAM using landmarks that are appropriate for its surrounding milieu. One thousand images of various indoor and outdoor environments were collected and used as training data for a three-layered feedforward backpropagation neural network. This neural network was then tested on two sets of query images of indoor environments and another two sets of outdoor environments, yielding 83% and 95% correct classification rates for the indoor images and 80% and 79% success rates for the outdoor images. © 2004 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Exploiting uniform vector instructions for GPGPU performance, energy efficiency, and opportunistic reliability enhancement 13

Exploiting uniform vector instructions for GPGPU performance...

引用

27th ACM International Conference on Supercomputing, ICS 2013

作者： Xiang, Ping Yang, Yi Mantor, Mike Rubin, Norm Hsu, Lisa R. Zhou, Huiyang Dept. of Electrical and Computer Engineering North Carolina State University Raleigh NC United States Dept. of Computing Systems Architecture NEC Laboratories America Princeton NJ United States Graphics Products Group AMD Inc. Orlando FL United States

ISBN: (纸本)9781450321303

State-of-art graphics processing units (GPUs) employ the single-instruction multiple-data (SIMD) style execution to achieve both high computational throughput and energy efficiency. As previous works have shown, there exists significant computational redundancy in SIMD execution, where different execution lanes operate on the same operand values. Such value locality is referred to as uniform vectors. In this paper, we first show that besides redundancy within a uniform vector, different vectors can also have the identical values. Then, we propose detailed architecture designs to exploit both types of redundancy. For redundancy within a uniform vector, we propose to either extend the vector register file with token bits or add a separate small scalar register file to eliminate redundant computations as well as redundant data storage. For redundancy across different uniform vectors, we adopt instruction reuse, proposed originally for CPU architectures, to detect and eliminate redundancy. The elimination of redundant computations and data storage leads to both significant energy savings and performance improvement. Furthermore, we propose to leverage such redundancy to protect arithmetic-logic units (ALUs) and register files against hardware errors. Our detailed evaluation shows that our proposed design has low hardware overhead and achieves performance gains, up to 23.9% and 12.0% on average, along with energy savings, up to 24.8% and 12.6% on average, as well as a 21.1% and 14.1% protection coverage for ALUs and register files, respectively. © 2013 ACM.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

IDENTIFICATION AND CONTROL EXPERIMENTATION WITH A REMOTE PLATFORM OF AN AUTONOMOUS IN-SCALE FAST-FERRY MODEL

引用

IFAC Proceedings Volumes 2007年第17期40卷 277-282页

作者： F.J. Velasco E. Revestido E. López E. Moyano Dept. Electronics Technology & Systems Engineering and Automatic Control Univ. Cantabria Dept. Sciences and Navigation Technics Engines and Naval Architecture Univ. País Vasco (UPV/EHU) Dept. Applied Mathematics & Computing Sciences Univ. Cantabria

In this paper a parametric lineal model of heading is obtained, using a remote experimentation platform for marine vehicles. This platform has an in-scale fast-ferry physical model, TF-120 which is autonomous and has all the elements required to emulate a real vessel. The physical model is controlled remotely from a PC using Wi-Fi communications. The tests for the data acquisition with the platform for marine vehicles were carried out in the surroundings of the Bay of Santander. The parametric model identified is used to design a PID controller for heading autopilot and a simulation of this system is performed in Simulink.

关键词： Remote marine laboratory (RML) parametric identification PID control heading control genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Introducing SLAMBench, a performance and accuracy benchmarking methodology for SLAM

Introducing SLAMBench, a performance and accuracy benchmarki...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Luigi Nardi Bruno Bodin M. Zeeshan Zia John Mawer Andy Nisbet Paul H. J. Kelly Andrew J. Davison Mikel Luján Michael F. P. O'Boyle Graham Riley Nigel Topham Steve Furber Dept. Computing Imperial College London UK Institute for Computing Systems Architecture The Univ. of Edinburgh Scotland School of Computer Science Univ. of Manchester UK

ISBN: (纸本)9781479969241

关键词： Three-dimensional displays Accuracy Benchmark testing Simultaneous localization and mapping Graphics processing units Kernel Cameras

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：