History repeats itself Since the invention of the programmable computer numerous computer scientists keep dedicating their professional lives to the design of "the single, best" programming model, whereas pr...
详细信息
ISBN:
(纸本)0769526438
History repeats itself Since the invention of the programmable computer numerous computer scientists keep dedicating their professional lives to the design of "the single, best" programming model, whereas programmers vote" by choosing their favourite languages and tools. Interestingly, these choices have always been guided by non-functional properties. For programming a single computer, the most widely used models have become object-oriented and component-based programming, a choice driven by their high abstraction level, leading to high programmer productivity For parallel computers, the winner turned out to be message passing, providing by far not the highest-possible abstraction level, but the closest match between machine architecture and programming model, leading to efficient program execution. For grids, the race is still open. Here, additional nonfunctional properties like fault-tolerance, security, and platform independence enter the scene. In this paper, we explore the scope of grid programming problems and argue for a palette of programming abstractions, each suitable for its respective problem domain.
This paper presents an analytical model for CMOS logic propagation delay which includes the effect of power supply noise. Using the nth power law model of MOSFETs, two scenarios are addressed: self-induced power suppl...
详细信息
ISBN:
(纸本)9780780393899
This paper presents an analytical model for CMOS logic propagation delay which includes the effect of power supply noise. Using the nth power law model of MOSFETs, two scenarios are addressed: self-induced power supply noise and globally-induced power supply noise. The analytical model is verified in simulation for both cases. The self-induced noise model matches simulation to within 0.36%. The globally-induced noise model matches simulation to within 5% for typical input rise time values and never more than 15% under extreme conditions.
(1)We develop an efficient linear programming detector (LPD) for multiple-input multiple-output (MIMO) systems. Instead of using the usual l(2) norm, our proposed LPD uses the l(1) norm as the detection metric, result...
详细信息
ISBN:
(纸本)9781424405053
(1)We develop an efficient linear programming detector (LPD) for multiple-input multiple-output (MIMO) systems. Instead of using the usual l(2) norm, our proposed LPD uses the l(1) norm as the detection metric, resulting in a mixed-integer linear program (MILP). Two branch-and-bound algorithms are proposed to solve the MILP. The solution of the MILP achieves the same full diversity order as the maximum likelihood detector. The MILP is further relaxed to a linear program (LP), which can be readily solved using the standard simplex method. We show that in some cases the solution of the LP is guaranteed to be that of the MILP. The LPD is also extended to the joint detection and decoding of linear block coded MIMO systems. Our LPD can be immediately implemented using mature circuits design for the simplex algorithm.
This paper presents the application of the Self-Adaptive Evolutionary programming optimization technique in the Power System State Estimation problem. State Estimation is a statistical-based optimization process used ...
详细信息
ISBN:
(纸本)9781424401727
This paper presents the application of the Self-Adaptive Evolutionary programming optimization technique in the Power System State Estimation problem. State Estimation is a statistical-based optimization process used in power systems to obtain the complex bus voltages of the system under study. The number of decision variables in the State Estimation problem grows in proportion to the size of the power system being considered. Therefore, it would be desirable to have robust state estimation procedures, with the ability to overcome premature stagnation or convergence problems. The Self-Adaptive Evolutionary programming method is an enhanced version of the canonical Evolutionary programming algorithm, which has been used successfully for solving high-dimensional continuous optimization problems. In order to validate the effectiveness of the proposed method, two case studies were considered. In the first one, the State Estimation problem was stated according with the Weighted Least Squares formulation and the results obtained on a 6-bus system were compared to those obtained via Newton's method. In the second case study, which was based on a 30-bus system, the results given by the Weighed Least Squares and Weighed Least Absolute Value formulations were compared to each other. The results from both case studies demonstrate the applicability of the proposed method for State Estimation purposes.
Genetic Parallel programming (GPP) evolves parallel programs for MIMD architectures with multiple arithmetic/logic processors (MAPs). This paper describes a tool intended for rapid development of GPP applications. A n...
详细信息
ISBN:
(纸本)1424401844
Genetic Parallel programming (GPP) evolves parallel programs for MIMD architectures with multiple arithmetic/logic processors (MAPs). This paper describes a tool intended for rapid development of GPP applications. A new software tool is proposed which is able to generate a simulator (in C language) of the MAP and a VHDL implementation of the MAP whose structure and parameters are specified in an input xml file. The proposed tool is intended to serve as first version of the core generator for MAPs utilized in GPP. Typical MAPs are synthetized and their performance is compared against the simulation running on a common PC for a typical task-a symbolic regression.
In this paper, we propose a new stereo matching technique which employs an adaptive multi-path dynamic programming (DP) scheme. DP has been a classical and popular optimization method for various computer vision probl...
详细信息
ISBN:
(纸本)9780780397323
In this paper, we propose a new stereo matching technique which employs an adaptive multi-path dynamic programming (DP) scheme. DP has been a classical and popular optimization method for various computer vision problems including stereo matching. However, the performance of conventional DP has not been satisfactory when applied to the stereo matching problems since the vertical correlation between scanlines has not been properly considered. This paper estimates disparity maps by using an optimal multi-directional DP scheme. We define a new energy function which considers the discontinuity of disparity and occlusions based on the edge information. The experimental results applied to the Middlebury stereo images [10] demonstrate that our proposed algorithm shows better performances in stereo matching than the previous DP based approaches.
We present a framework for a parallel programming model by remote procedure calls bridging between large-scale computing resource pools managed by multiple grid-enabled-job scheduling systems. With this system, the us...
详细信息
ISBN:
(纸本)0769525857
We present a framework for a parallel programming model by remote procedure calls bridging between large-scale computing resource pools managed by multiple grid-enabled-job scheduling systems. With this system, the user can exploit not only each remote servers and clusters, but also computing resources provided with grid-enabled job scheduling systems located on different sites. This framework requires a Grid RPC system to decouple the computation in a remote node from the Grid RPC mechanism and uses document-based communication rather than connection-based communication. We implemented the proposed framework as an extension of the OmniRPC system, which is a Grid RPC system for parallel programming in a grid environment. We designed a general interface to adapt the OmniRPC system to various grid-enabled job scheduling systems easily and applied the proposed system to several grid-enabled job scheduling systems, including XtremWeb, CyberGRIP, Condor and Grid Engine. we show the preliminary performance of these implementations using a phylogenetic application. We found that the proposed system can achieve approximately the same performance as using OmniRPC and can handle interruptions in worker programs on remote nodes.
According to the optimization control modes of performance seeking control (PSC) in aeroengine, the aeroengine performance optimization problems which were described as a linear programming problem are studied. In res...
详细信息
ISBN:
(纸本)0780393953
According to the optimization control modes of performance seeking control (PSC) in aeroengine, the aeroengine performance optimization problems which were described as a linear programming problem are studied. In respect that the linear programming is only guaranteed to converge to a local minimized and nonlinear method is much more computation, a hybrid algorithm based on linear programming (LP) and Model-Assisted Pattern Search (MAPS) is provided in this paper. And the hybrid optimization method is used for aeroengine control. Digital simulations are conduced in maximum thrust mode and minimum fuel mode in this paper. Simulation results demonstrate that the engine performance can be improved further and calculation time can be reduced rapidly by using hybrid algorithm.
Establishing suitable programming models for pervasive spaces is essential in improving the productivity, enhancing the quality of pervasive systems, and creating an open platform for interoperability. Two different m...
详细信息
ISBN:
(纸本)0769525105
Establishing suitable programming models for pervasive spaces is essential in improving the productivity, enhancing the quality of pervasive systems, and creating an open platform for interoperability. Two different models, namely, the context-driven model and the service-oriented model, have been proposed and studied for their feasibilities as the foundation for implementing programmable pervasive spaces. We present these two alternatives and contrast their advantages and disadvantages.
A novel technique is proposed to achieve highly linear programmable voltage-to-current (V-I) converters over a rail-to-rail input range. It employs two passive resistors with one of their terminals tied to a very low ...
详细信息
ISBN:
(纸本)9780780393899
A novel technique is proposed to achieve highly linear programmable voltage-to-current (V-I) converters over a rail-to-rail input range. It employs two passive resistors with one of their terminals tied to a very low impedance node to achieve an almost distortionless V-I conversion for rail-to-rail input signals. The resistor currents are conveyed to the output via programmable floating-gate MOS current mirrors that work in moderate inversion, thus achieving transconductance tuning, in a wide range while preserving linearity in the rail-to-rail input range. Initial charge of the floating gates is removed during fabrication. An implementation fabricated in a 0.5-mu m CMOS technology using a dual supply of +/- 1.5 V achieves a THD of -60dB over a wide transconductance tuning range with differential input swings equal to the total supply voltage. The area and power consumption are 0.1 mm(2) and 2.6 mW.
暂无评论