检索结果-内蒙古大学图书馆

IEEE International Conference on computer Design: VLSI in computers and Processors, (ICCD)

作者： J.-P. Theis L. Thiele Applied Microelectronics Laboratory Public Research Center Henri Tudor Luxembourg Luxembourg Computer Engineering and Networks Laboratory Swiss Federal Institute of Technology Zurich Switzerland

This paper deals with the cyclo-static (periodic) scheduling of iterative instruction sequences on VLIW-processor architectures, in the case when the period is imposed (periodic real time constraints). This problem naturally arises when designing VLIW-processor architectures as I/O interfaces between cyclo-static-processor arrays and the external world. Three important mapping problems related to periodic real time constraints are discussed, depending on the VLIW-processor resource constraints and optimization criteria that are considered. Precise conditions under which there exist cycle-static schedules are formulated. Whenever possible, we present systematic and optimal mapping methodologies with polynomial solution time. The starting point is thereby a novel description of iterative instruction sequences in the form of generalized directed acyclic graphs (DAG's). Finally, an innovative VLIW-processor architecture, called Extended Periodic Operation Model (EPOM) with distributed shift register file, which optimally fits into the mapping trajectory, is presented.

关键词： Clocks Processor scheduling Polynomials Flow graphs Hardware Coprocessors Microelectronics Shift registers Iterative methods Optimal control

来源：评论

学校读者我要写书评

暂无评论

Synthesis of domain-specific heterogeneous multiprocessor systems: hybrid video coding schemes

Synthesis of domain-specific heterogeneous multiprocessor sy...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： M. Schobinger L. Thiele Corporate Research & Development Siemens AG Munich Germany Computer Engineering and Networks.Laboratory Swiss Federal Institute of Technology Zurich Switzerland

The purpose of the paper is to describe the system SYLT (synthesis using integer linear programming techniques) for design space exploration for flexible, programmable heterogeneous multiprocessor architectures. In contrast to comparable attempts, the synthesis system is dedicated to DSP applications and takes into account computation, communication and memory constraints. Models, methods and application are explained in detail by means of a comprehensive case study.

关键词： Multiprocessing systems Video coding computer architecture Digital signal processing Network synthesis Costs Signal processing algorithms Processor scheduling Memory management Energy consumption

来源：评论

学校读者我要写书评

暂无评论

On minimax detection of signals with unknown parameters

On minimax detection of signals with unknown parameters

引用

Asilomar Conference on Signals, Systems & computers

作者： G.A. Hanson R.A. Iltis Weapons Division Naval Air Warfare Center China Lake CA USA Communication Networks Laboratory Department of Electrical and Computer Engineering University of California Santa Barbara CA USA

We review two different strategies for minimax detection of complex signals with unknown parameters in additive Gaussian noise. The first strategy is based on the least-favorable distribution of the unknown parameters. The second strategy is a distribution-free approach which seeks to maximize performance for the worst case parameter values. Conditions are given under which the two detectors produce identical performance.

关键词： Minimax techniques Signal detection Detectors Matched filters Signal to noise ratio Robustness Uncertainty Weapons Communication networks Laboratories

来源：评论

学校读者我要写书评

暂无评论

Comparison of a new hybrid connectionist-SCHMM approach with other hybrid approaches for speech recognition

Comparison of a new hybrid connectionist-SCHMM approach with...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： H.-P. Hutter Speech Processing Group Computer Engineering and Networks Laboratory ETH-Zentrum Zurich Switzerland

The paper compares a newly proposed hybrid connectionist-SCHMM approach [Hutter and Pfister 1994] with other hybrid approaches. In the new approach a multilayer perceptron (MLP) replaces the conventional codebooks of semi-continuous HMMs. The MLP is therefore trained on so-called basic elements (phones and phone parts) in such a way that the outputs of the network estimate the a posteriori probabilities of these elements, given a context of input vectors. These a posteriori estimates are converted into scaled likelihoods, which are then used as observation probabilities in the framework of classical SCHMMs. The remaining parameters of the SCHMMs are trained with the well-known Baum-Welch algorithm using the estimated likelihoods of the MLP. This approach compared favorably with other proposed hybrid systems and classical approaches on an isolated German digit recognition task over telephone lines. It exhibited the highest recognition rate of all systems, followed by an approach using LVQ3 optimization of the codebook.

关键词： Hidden Markov models Probability Speech recognition Speech processing computer networks Laboratories Multilayer perceptrons Telephony Maximum likelihood estimation Costs

来源：评论

学校读者我要写书评

暂无评论

Packet arrival scheduling at a exponential server for min-max blocking

Packet arrival scheduling at a exponential server for min-ma...

引用

IEEE International Conference on Communications (ICC)

作者： R. Yates C. Rose Wireless Information Networks Laboratory Department of Electrical and Computer Engineering Rutgers university Piscataway NJ USA

Consider a source using a high speed data link to transmit packets to a downstream buffer. It is argued that allocating finite buffer space to each source using the link encourages each source to regulate its packet arrival process to avoid packet blocking downstream. For bursty sources using high speed links, an entire burst might be transmitted to the downstream buffer before any feedback would be received. Regulating the packet transmission process to minimize packet blocking at the downstream link is formulated as a stochastic scheduling problem. For an exponential service facility with a service rate that is a nondecreasing function of queue occupancy, it is found that a type of threshold schedule minimizes the maximum packet blocking probability. Under light loading, these optimal threshold schedules exhibit a form of burstiness. In addition, these threshold schedules are then compared to a leaky bucket regulator.

关键词： Feedback Stochastic processes Network servers Propagation delay Processor scheduling Laboratories computer networks Data engineering Regulators High-speed networks

来源：评论

学校读者我要写书评

暂无评论

Cross-validation without a validation set in BP-trained neural nets

Cross-validation without a validation set in BP-trained neur...

引用

International Conference on Neural networks

作者： M.H. Hassoun P.B. Watta R. Shringarpure The Computation and Neural Networks Laboratory Department of Electrical & Computer Engineering Wayne State University Detroit MI USA

Generalization in backprop-trained multilayer neural networks is discussed for problems where training data is either scarce or else extremely costly to obtain. In this case, the usual method of cross-validation, whereby the data set is partitioned into training, testing, and validation sets, is not feasible. In this paper we demonstrate that it is sometimes possible to use all available data for training a large network (a network capable of overfitting the data) and yet still determine an appropriate stopping point to ensure that the network generalizes properly.

关键词： Neural networks Multi-layer neural network Training data Testing

来源：评论

学校读者我要写书评

暂无评论

POM: a processor model for image processing

POM: a processor model for image processing

引用

IEEE International Conference on computer Design: VLSI in computers and Processors, (ICCD)

作者： J.-P. Theis L. Thiele Applied Microelectronics Laboratory Public Research Center Henri Tudor Luxembourg Computer Engineering and Networks Laboratory Swiss Federal Institute of Technology Zurich Switzerland

In this paper, we describe a new processor model called Periodic Operation Model (POM) that is suitable for real time image processing. First we analyze existing image processing systems in order to situate our approach. Starting from the processor architecture, we derive the corresponding algorithm class by means of a novel hardware description. Then we address the allocation and scheduling problem. We show that allocation and scheduling can be decoupled in the mapping process related to POM-processor arrays and outline the principles of an optimal mapping trajectory. We describe the outline of a novel ILP-model for allocation of POM-processor arrays which takes into account array-topology and bus bandwidth constraints. Finally we discuss implementational aspects of the POM as well as applications in image processing. We especially show that POM-processor arrays can be integrated onto single chips, thereby allowing to achieve several GOPS processing power per chip.

关键词： Image processing Complexity theory Clocks Costs Laboratories Processor scheduling Application software Shift registers Microelectronics Image analysis

来源：评论

学校读者我要写书评

暂无评论

OPTIMUM FREQUENCY REUSE IN MOBILE TELEPHONY SYSTEMS

引用

INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS 1995年第3期8卷 185-190页

作者： KAZANTZAKIS, MG DEMESTICHAS, PP ANAGNOSTOU, ME National Technical University of Athens Computer Science Division Department of Electrical and Computer Engineering Heroon Polytechneiou 9 Zografou Athens 15773 Greece Michael G. Kazantzakii was born in Athens Greece in 1962. He received both the Diploma and Ph.D. Degree from the Electrical Engineering Department of the National Technical University of Athens in 1986 and 1992 respectively. During the period 1992–1993 he was a member of the Staff of the Hellenic Navy Research Centre working in the area of underwater acoustics. Since 1994 he has been a researcher at the National Technical University of Athens Department of Electrical Engineering and Computer Science Telecommunications Laboratory. His interests include data mobile and integrated services networks. Panagiotis P. Demestichas was born in Athens Greece in 1967. He received the Diploma in electrical engineering from the National Technical University of Athens (NTUA) Athens in 1993. At present he is a research assistant at NTUA working towards his Ph.D. dissertation. He participated in the BRITE/EURAM-EURATN and in the RACE-TOMQAT and MONET projects. His research interests are in the areas of broadband and mobile networks queueing theory algorithms and complexity theory. He is a Member of the Technical Chamber of Greece. Miltiades E. Anagnostou was born in Athens in 1958. He received the Electrical Engineers Diploma from the National Technical University of Athens (NTUA) in 1981. During the period 1981–1987 he was a teaching assistant at NTUA. In 1987 he received his Ph.D. in the area of computer networks. In 1987–1989 he served the Greek army in the Air Force Research and Development Center. In 1989–1991 he was a Lecturer and since 1991 he has been an Assistant Professor in the Computer Science Division Department of Electrical and Computer Engineering of the NTUA. He teaches courses on modern telecommunications formal specification of protocols stochastic processes and queueing theory. Since 1981 Dr. Anagnostou has been involved

The reason why cellular mobile telephony systems exist is that they allow frequency reuse. Dynamic allocation algorithms improve the network efficiency and the service quality. Dynamic allocation is closely related to the following instantaneous allocation problem: given a number of channels (frequencies), a cell structure and the number of calls to be accommodated in each cell, find the optimum allocation of channels to cells subject to restrictions concerning the distance of cells, where the same frequency can be reused. In this paper we formulate and solve this problem by showing that it can be reduced to a 0-1 programming problem. Finally we present results, and draw subsequent conclusions.

关键词： MOBILE TELEPHONY FREQUENCY ALLOCATION UMTS

来源：评论

学校读者我要写书评

暂无评论

Evaluation of a minimum power handoff algorithm

Evaluation of a minimum power handoff algorithm

引用

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

作者： Chen-Nee Chuah R.D. Yates Wireless Information Networks Laboratory (WINLAB)Department of Electrical and Computer Engineering Rutgers University Piscataway NJ USA Rutgers The State University of New Jersey New Brunswick NJ US Dept. of Electr. & Comput. Eng. Rutgers Univ. Piscataway NJ USA

Previous work has shown that a handoff algorithm based on SIR alone (SIR based handoff) prohibits handoffs near nominal cell boundaries, causing cell dragging and unnecessarily high transmitter powers. We propose the minimum power handoff (MPH) algorithm in which mobiles constantly search for a combination of base and channel assignment that minimizes the uplink transmitted power. This algorithm is shown to reduce the average power level by 4dB compared to SIR based handoff. Results show that using received power as a handoff criterion reduces call dropping but increases the number of unnecessary handoffs significantly. To avoid a "ping pong" effect, a timer is introduced to delay the intercell handoff.

关键词： Base stations Power control Resource management Radio transmitters Fading Interference Laboratories computer networks Power engineering computing Power engineering and energy

来源：评论

学校读者我要写书评

暂无评论

A platform for multimedia information exchange 94

A platform for multimedia information exchange

引用

1994 ACM Symposium on Applied Computing, SAC 1994

作者： Huang, Chung-Ming Lo, Chung-Ming Laboratory of Computer Networks and Multimedia Systems Institute of Information Engineering National Cheng- Kung University Tainan70101 Taiwan

ISBN: (纸本)0897916476

This paper presents a computer-supported desk-co- desk information exchange system which allows two users to conduct a multimedia information transfer. Participants can jointly view and edit a piece of shared multimedia information, including text, graphics, images, and voice, which is transferred through a local area network. The system includes a talk system combined with voice and text, a co-design system, including a co-draw component and a co-editor com-ponent, a document transfer system and a multimedia e-mail system. The system is based on TCP and UDP protocols and is developed on the SUN SPARC OPENLOOK system. This system is applicable to a number of environments, e.g., office automation, inter-hospital network service, computer supported collaboration work, distributed software development, group decision support, and so on. © 1994 ACM.

关键词： Decision support systems

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：