检索结果-内蒙古大学图书馆

International Conference on High Performance computing in the Asia-Pacific Region

作者： Meng-Shiou Wu R.A. Kendall S. Aluru Department of Electrical and Computer Engineering Scalable Computing Laboratory US DOE Ames IA USA Department of Electrical and Computer Engineering Department of Computer Science Ames IA

We investigate two strategies to improve the performance of collective communications on a cluster of SMPs. In the first part of the paper, we explore the advantages and disadvantages of the commonly used strategy "optimize shared memory send and receive only". In the second part of the paper, we show a model to design collective communications on SMP clusters which can be used to improve the disadvantages of the former strategy. Several collective operations are implemented through the later process, and the experimental results show better performance than vendor supplied implementations.

关键词： Switched-mode power supply Laboratories Clustering algorithms US department of Energy computer science Optimization methods Broadcasting Libraries Testing Communication standards

来源：评论

学校读者我要写书评

暂无评论

Cancellation of Unwanted Audio to Support Interactive computer Music 30

Cancellation of Unwanted Audio to Support Interactive Comput...

引用

30th International computer Music Conference, ICMC 2004

作者： Lee, Jonghyun Dannenberg, Roger B. Chun, Joohwan Scientific Computing Laboratory Department of Electrical Engineering Korea Advanced Institute of Science and Technology 373-1 Kusong-Dong Yusong-Gu Taejeon305-701 Korea Republic of School of Computer Science Carnegie Mellon University United States

A real-time unwanted-audio cancellation system is developed. The system enhances recorded sound by canceling unwanted loudspeaker sounds picked up during the recording. After cancellation, the resulting sound gives an improved estimation of the live performer’s sound. The cancellation works by estimating the unwanted audio signal and subtracting it from the recorded signal. The canceller is composed of a delay block and two adaptive digital filters. Our work extends conventional echo-cancellation methods to address problems we encountered in music applications. We describe a real-time implementation in Aura and present experimental results in which the proposed canceller enhances the performance of a real-time pitch detector. The cancellation ratio is measured and limitations of the system are discussed. © ICMC 2004. All rights reserved.

关键词： Digital filters

来源：评论

学校读者我要写书评

暂无评论

Carrier frequency recovery in MIMO and OFDM systems using orthogonal training sequences

Carrier frequency recovery in MIMO and OFDM systems using or...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Kyunchun Lee Joohwan Chun Scientific Computing Laboratory Department of Electrical Engineering and Computer Science Korea Advanced Institute of Science and Technology Daejeon South Korea

ISBN: (纸本)0780385217

We propose a training-sequence-based frequency offset estimator for a multiple transmit-and-receive antenna system in frequency-flat fading channels. The estimator is based on a maximum likelihood (ML) criterion and does not require channel information. To reduce the computational load, we propose to use special training sequences - the periodic orthogonal codes. Using these codes, we get a closed form estimator which requires much lower computational load (some additions and multiplications). For a high signal-to-noise ratio (SNR) and small frequency offset, the proposed estimator achieves the performance of the optimal ML estimator, which locates the peak of the likelihood function. We also apply the proposed estimator to an orthogonal frequency division multiplexing (OFDM) system. With simulations, we test the performance of the proposed estimator.

关键词： MIMO OFDM Frequency estimation Transmitting antennas Maximum likelihood estimation Receiving antennas Fading Oscillators AWGN Scientific computing

来源：评论

学校读者我要写书评

暂无评论

On the beamforming weight of the MIMO/SDMA system under channel uncertainty

On the beamforming weight of the MIMO/SDMA system under chan...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Kyungchun Lee Joohwan Chun Scientific Computing Laboratory Department of Electrical Engineering and Computer Science Korea Advanced Institute of Science and Technology Daejeon South Korea

ISBN: (纸本)0780382552

We consider beamforming methods for the downlink of space division multiple access (SDMA) systems when there are channel information errors. A single antenna at the mobile is generally assumed in SDMA systems, but we consider multi-input and multi-output (MIMO) systems which adopt multiple antennas at both transmitter and receiver. If we know the exact channel information of each user, cochannel interference among users can be perfectly removed by the zero-forcing nulling method in MIMO/SDMA systems. However, the channel estimate always contains estimation errors in real systems, which can cause significant performance degradation. To cope with channel information errors, we develop two beamforming weights based on null-space constraints. The first one is designed to minimize the transmit power while not influencing the desired receive signal. The interfering effect from channel information errors is reduced since the transmit power of cochannel interfering users is minimized. The second one is designed to minimize the power of the received interference signal. It computes the beamforming weight which has the minimum expectation of interference power with knowledge of the variance of channel estimation errors. The proposed beamforming weights are tested by computer simulations.

关键词： Array signal processing MIMO Multiaccess communication Uncertainty Mobile antennas Receiving antennas Transmitting antennas Signal design Interference Downlink

来源：评论

学校读者我要写书评

暂无评论

Coordinated aggregate scheduling for improving end-to-end delay performance

Coordinated aggregate scheduling for improving end-to-end de...

引用

International Workshop on Quality of Service

作者： W. Sun K.G. Shin Real-Time Computing Laboratory Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI USA

This paper proposes a novel coordinated aggregate scheduling (CAS) algorithm that combines both EOF (earliest-deadline-first) scheduling and rate-based fair queueing. CAS uses guaranteed rate (GR) scheduling (Goyal, P et al., 1995) for traffic aggregates at the inter-aggregate level, but employs EDF-like scheduling at the intra-aggregate level. Computation of the deadline D/sub N/ of a packet at an intermediate node N is coordinated between the node N and its upstream nodes, and D/sub N/ is related to the packet's guaranteed rate clock (GRC) value at the flow-aggregation node. CAS provides tighter end-to-end (e2e) delay bounds than the "vanilla" GR aggregate scheduling that relies on FIFO queueing within an aggregate. Our in-depth simulation results demonstrate CAS's superior performance. Moreover, as an aggregate-based work-conserving scheduling algorithm, CAS incurs lower scheduling and state-maintenance overheads at routers than per-flow scheduling. These salient features make CAS very attractive for use in Internet core networks.

关键词： Aggregates Delay Scheduling algorithm Traffic control Processor scheduling Content addressable storage Telecommunication traffic Clocks IP networks Multiprotocol label switching

来源：评论

学校读者我要写书评

暂无评论

Component allocation with multiple resource constraints for large embedded real-time software design

Component allocation with multiple resource constraints for ...

引用

IEEE Symposium on Real-Time and Embedded Technology and Applications

作者： S. Wang J.R. Merrick K.G. Shin Real-Time Computing Laboratory Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI USA

Allocating software components while meeting multiple platform resource constraints is crucial for model-based design of large embedded real-time software and automatic design model transformation. In this paper, we propose a new method for component allocation using an informed branch-and-bound and forward checking mechanism subject to a combination of resource constraints. We have implemented this method in the automatic integration of reusable embedded software (AIRES) toolkit - which has been developed under the DARPA MoBIES Program - and applied it to an automotive electronic throttle control (ETC) system. Our evaluation based on randomly-generated design models has shown that the proposed method scales well for large and complex embedded real-time software.

关键词： Resource management Software design Embedded software Control systems Embedded computing Forward contracts Automotive electronics Automatic control Software performance Scheduling

来源：评论

学校读者我要写书评

暂无评论

On extended finite element method (XFEM) for modelling of organ deformations associated with surgical cuts

引用

International Symposium on medical Simulation, ISMS 2004

作者： Vigneron, Lara M. Verly, Jacques G. Warfield, Simon K. Signal Processing Group Department of Electrical Engineering and Computer Science University of Liège Belgium Surgical Planning Laboratory Brigham and Women’s Hospital and Harvard Medical School Boston United States

ISBN: (纸本)3540221867

The Extended Finite Element Method (XFEM) is a technique used in fracture mechanics to predict how objects deform as cracks form and propagate through them. Here, we propose the use of XFEM to model the deformations resulting from cutting through organ tissues. We show that XFEM has the potential for being the technique of choice for modelling tissue retraction and resection during surgery. Candidates applications are surgical simulators and image-guided surgery. A key feature of XFEM is that material discontinuities through FEM meshes can be handled without mesh adaptation or remeshing, as would be required in regular FEM. As a preliminary illustration, we show the result of XFEM calculation for a simple 2D shape in which a linear cut was made. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Algorithmic-Parameter optimization of a parallelized split-step fourier transform using a modified BSP cost model 04

引用

Proceedings of the Second international conference on Parallel and Distributed Processing and Applications

作者： Elankovan Sundararajan Malin Premaratne Shanika Karunasekera Aaron Harwood Department of Computer Science and Software Engineering The University of Melbourne Carlton Victoria Australia Advanced Computing and Simulation Laboratory Department for Electrical and Computer System Engineering Monash University Clayton Victoria Australia

ISBN: (纸本)9783540241287

Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Level set methods in an EM framework for shape classification and estimation

引用

medical Image computing and computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings

作者： Tsai, Andy Wells, William Warfield, Simon K. Willsky, Alan Department of Medicine Massachusetts General Hospital Harvard Medical School Boston MA United States Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge MA United States Department of Radiology Brigham and Women's Hospital Harvard Medical School Boston MA United States Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology Cambridge MA United States Department of Radiology Children's Hospital Harvard Medical School Boston MA United States

ISBN: (纸本)3540229760

In this paper, we propose an Expectation-Maximization (EM) approach to separate a shape database into different shape classes, while simultaneously estimating the shape contours that best exemplify each of the different shape classes. We begin our formulation by employing the level set function as the shape descriptor. Next, for each shape class we assume that there exists an unknown underlying level set function whose zero level set describes the contour that best represents the shapes within that shape class. The level set function for each example shape is modeled as a noisy measurement of the appropriate shape class's unknown underlying level set function. Based on this measurement model and the judicious introduction of the class labels as hidden data, our EM formulation calculates the labels for shape classification and estimates the shape contours that best typify the different shape classes. This resulting iterative algorithm is computationally efficient, simple, and accurate. We demonstrate the utility and performance of this algorithm by applying it to two medical applications. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Maximum principle

来源：评论

学校读者我要写书评

暂无评论

Sequential and Parallel Meta-Heuristics for Solving the Single Row Routing Problem

引用

Cluster computing 2004年第2期7卷 123-139页

作者： Zomaya, Albert Y. Patterson, Daniel Olariu, Stephan School of Information Technologies The University of Sydney Sydney Australia Parallel Computing Research Laboratory Department of Electrical and Electronic Engineering The University of Western Australia Australia Department of Computer Science Old Dominion University Norfolk USA

The design of multi-layer printed circuit boards is vital in the construction of complex electronic systems. Wire routing is a crucial step in the overall design process, which can be decomposed into a number of single row routing (SRR) problems. This paper proposes an approach to solve the SRR problem based on parallel meta-heuristics. The development of this technique involves the design of an encoding strategy that allows all possible routings to be uniquely represented and the derivation of cost functions that maximizes the quality of the developed solutions. Further, parallelization of the proposed approach is attempted to improve the computational efficiency. The different stages of the development are backed by experiments to show the pros and cons of the sequential and parallel implementations.

关键词：

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：