Search

In this paper, we present a new filter model which combines Gibbs random field with anisotropic-diffusion. The Gibbs random field is used to determine the boundaries of the objects in image according to the spatial information of image. Then the anisotropic-diffusion propagates different energy at different orientation with respect to conduction coefficient, and stops diffusing at the boundaries of the objects in image. We also provide the numerical implementation of the proposed method. The numerical experimental results show that our method has a high performance. [ABSTRACT FROM AUTHOR]

Local controller and remote learning device are connected through communication network in Networked Learning based Control Systems (NLCSs). Network-induced delays are inevitable during data transmission, and will deteriorate the real-time transmission of learning result from learning device to controller and even destabilize the entire system. This paper deals with the delays in NLCSs. The sources of delays are discussed and the possibility of reducing all delays into an equivalent delay is proposed. Finally, experimental measurements for that equivalent delay on Ethernet are conducted. [ABSTRACT FROM AUTHOR]

This paper presents a high-level, algorithmic, single-assignment programming language and its tailor-made optimizing compiler. The compiler is able to generate a synthesizable hardware description language for reconfigurable systems based on input instruction set. Simulated annealing and force-directed scheduling approaches were employed in this compiler for design speed and resource optimizations. The tasks of the designed compiler include control flow graph transformation, component selection, component scheduling, and VHDL transformation. Language features are introduced and the structure of the compiler is discussed. [ABSTRACT FROM AUTHOR]

Shelling is a powerful modeling operation in current CAD /CAM systems. In this paper, we present a new shelling algorithm for regular B-Rep solids. We first generate an initial offset solid, then correct the self-intersecting offset solid if necessary. We obtain the resultant shelling solid through a regularized Boolean operation finally. The algorithm has been implemented in a solid modeling system, and some examples are given to illustrate it. [ABSTRACT FROM AUTHOR]

Although intensive work has been done in the area of load balancing, the grid computing environment is different from the traditional parallel systems, which prevents existing load balancing schemes from benefiting large-scale parallel applications. This paper provides a survey of the existing solutions and new efforts in load balancing to address the new challenges in grid computing. We classify the surveyed approaches into three categories: resource-aware repartition, divisible load theory and prediction based schemes. [ABSTRACT FROM AUTHOR]

In this paper, Adaptive Modulation and Coding (AMC) is combined with Multiple Input Multiple Output (MIMO) multiplexing to improve the throughput performance of AMC. In addition, a system that adopts Selection Transmit Diversity (STD) in the AMC-MIMO multiplexing system is proposed. The received SNR is improved by adopting STD techniques. And it increases probability of selecting MCS (Modulation and Coding Scheme) level that supports higher data rate. This leads to an increased throughput of the AMC-MIMO multiplexing system. STD in our simulation selects 2 transmission antennas from 4 antennas and AMC-MIMO multiplexing process operates with the selected antennas. The computer simulation is performed in flat Rayleigh fading channel. The results show that the proposed system achieves a gain of 1Mbps over the AMC-MIMO multiplexing system with the same number of antennas at 15dB SNR. [ABSTRACT FROM AUTHOR]

We propose a new type of preconditioners for symmetric Toeplitz system Tx = b . When applying iterative methods to solve linear system with matrix T, we often use some preconditioner C by the preconditioned conjugate gradient (PCG) method [3]. If T is a symmetric positive definite Toeplitz matrix, two kinds of preconditioners are investigated: the "optimal" one , which minimizes C - T F, and the "superoptimal" one , which minimize I - C-1T F [8]. In this paper, we present a general approach to the design of Toeplitz preconditioners based on the optimal investigating and also preconditioners C with preserving the characteristic of the given matrix T. Fast all resulting preconditioners can be inverted via fast transform algorithms with O(N log N) operations. For a wide class of problems, PCG method converges in a finite number of iterations independent of N so that the computational complexity for solving these Toeplitz systems is O(N log N ) [2]. [ABSTRACT FROM AUTHOR]

A physics engine in computer games takes charge of the calculations simulating the physical world. In this paper, we evaluate the performance of three numerical integral methods: Euler method, Improved Euler method, and Runge-Kutta method. We utilized a car moving game for the simulation experiments logging fps (frame per second). Each numerical integral was evaluated under two different settings, one with collision detection and the other without it. The simulation environment without collision detection was divided into two sections, a uniform velocity section and a variable velocity section. The Euler method was shown to have the best fps in the simulation environment with collision detection. Simulation with collision detection shows similar fps for all three methods and the Runge-Kutta method showed the greatest accuracy. Since we tested with rigid bodies only, we are currently studying efficient numerical integral methods for soft body objects. [ABSTRACT FROM AUTHOR]

In order to use parallel computers in specific applications, algorithms need to be developed and mapped onto parallel computer architectures. Main memory access for shared memory system or global communication in message passing system deteriorate the computation speed. In this paper, it is found that the m-step generalization of the block Lanczos method enhances parallel properties by forming m simultaneous search direction vector blocks. QR factorization, which lowers the speed on parallel computers, is not necessary in the m-step block Lanczos method. The m-step method has the minimized synchronization points, which resulted in the minimized global communications and main memory accesses compared to the standard method. [ABSTRACT FROM AUTHOR]

Double-loop and 2-circulant networks are widely used in the design and implementation of local area networks and parallel processing architectures. In this paper, we investigate the routing of a message on circulant networks, that is a key to the performance of this network. We would like to transmit 2k packets from a source node to a destination node simultaneously along paths on G(n; ±s1,±s2, ...,±sk), where the ith packet will traverse along the ith path (1≤ i ≤2k). In oder for all packets to arrive at the destination node quickly and securely, the ith path must be node-disjoint from all other paths. For construction of these paths, employing the Hamiltonian Circuit Latin Square(HCLS), a special class of (n × n) matrices, we present O(n2) parallel routing algorithm on circulant networks. [ABSTRACT FROM AUTHOR]

For finding accurate and biologically meaningful multiple sequence alignment, it is required to consider a loosely constrained version of multiple sequence alignment. This paper is the first to define a loosely constrained multiple sequence alignment (LCMSA). And it also designs a complete algorithm for the loosely constrained pair-wise sequence alignment problem with the time complexity of and an approximation algorithm for LCMSA running in time , k and n is the number and the maximum length of input sequences, and α is the length of constrained sequence. In fact, α is small compared to n and it is reasonable to regard it as a constant. So, the former two algorithms run in polynomial time. Finally, a LCMSA software system is built up. [ABSTRACT FROM AUTHOR]

The natural preconditioner defined by local potentials is effective in the truncated Newton method for minimizing large scale biomolecular potential energy functions. This paper extends its definition, and proposes an algorithm for generating the sparse pattern of the preconditioner from the primary structure of a molecular system with N atoms. It shows that the implementation of the new compressed sparse preconditioner requires only a linear order of N memory locations. [ABSTRACT FROM AUTHOR]

This paper presents a distributed genetic algorithm for the channel routing problem in MPI environments. This system is implemented on a network of personal computers running Linux operating system connected via 10Mbps Ethernet. Each slave processor generates its own sub-population using genetic operations and communicates with the master processor in an asynchronous manner to form the global population. The experimental results show that the proposed algorithm maintains the convergence properties of sequential genetic algorithm while it achieves linear speedup as the nets of the channel routing and the number of computing processors increase. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel approach to the load balancing problem, which is an important issue in parallel processing. The proposed load balancing algorithm called Mean Field Genetic Algorithm (MGA) is a hybrid algorithm of Mean Field Annealing (MFA) and Simulated annealing-like Genetic Algorithm (SGA). The proposed MGA combines the benefit of rapid convergence property of MFA and the effective genetic operations of SGA. Our experimental results indicate that the composition of heuristic mapping methods improves the performance over the conventional ones in terms of communication cost, load imbalance and maximum execution time. [ABSTRACT FROM AUTHOR]

In this paper, we propose a genetic algorithm(GA) based load balancing approach for a distributed system. Fisrt we develop two GA based algorithms to effectively reduce unnecessary request messages in sender-initiated approach and receiver-initiated approach respectively and then construct an adaptive algorithm for load balancing in a distributed system by combining the two GA based algorithms. The experiments show the proposed GA based adaptive algorithm performs efficiently in the variance of load of the distributed system. [ABSTRACT FROM AUTHOR]

This paper introduces the semantic of Multiparty Joint Authentication (MJA) into the authentication service, which is to find simplified or optimal authentication solutions for all involved principals as a whole instead of to authenticate each pair of principals in turn. MJA is designed to support multiparty security contexts in grids with a specified, understood level of confidence and reduce the time cost of mutual authentications. Graph theory model is employed to define MJA, and analyze its mathematical properties. Two algorithms to find an n-principal, n-order MJA solution are also presented and proved. [ABSTRACT FROM AUTHOR]

For the reason of all the centers and radii needed to be adjusted iteratively, the learning speed of radial basis function (RBF) neural networks is always far slower than required, which obviously forms a bottleneck in many applications. To overcome such problem, we propose a fast and accurate RBF neural network in this paper. First we prove the universal approximation theorem for RBF neural networks with arbitrary centers and radii. Based on this theory, we propose a new learning algorithm called fast and accurate RBF neural network with random kernels (RBF-RK). With the arbitrary centers and radii, our RBF-RK algorithm only needs to adjust the output weights. The experimental results, on function approximation and classification problems, show that the new algorithm not only runs much faster than traditional learning algorithms, but also produces better or at least comparable generalization performance. Keywords: RBF, random kernel, fast, accurate. [ABSTRACT FROM AUTHOR]

In this paper, we propose to exploit pre-established label switched paths with the aim to reduce handover processing latency which is one of the most critical performance metrics in micro-mobility environment. Thus, our proposed scheme has an advantage of offering micro-mobility to Mobile IP (MIP) with lower-latency handover as compared with the several existing MPLS-based MIP approaches. An analytic model of setting up LSPs for fast handover processing is presented. Further, we investigate the performance of our model. [ABSTRACT FROM AUTHOR]

This paper deals with a balanced model reduction for a class of Takagi-Sugeno(T-S) fuzzy systems with uncertain time varying parameters. We define a generalized controllability Gramian and a generalized observability Gramian for quadratically stable T-S fuzzy systems with uncertainties. We introduce a balanced state space realization using the generalized controllability and observability Gramians and obtain a reduced model by truncating not only states but also time varying uncertain parameters from the balanced state space realization. We also present an upper bound of the approximation error. The generalized controllability and observability Gramians can be computed from solutions of linear matrix inequalities(LMI's). [ABSTRACT FROM AUTHOR]

In this paper, we analyze the burst queue with burst assembler at an edge in optical burst-switched networks. It is significant from the fact that most of processing and buffering are concentrated at the edge in OBS networks. Modelling the burst queue in terms of IP packets, we aim to obtain the expected number of packets in burst queue the size of which influences the cost of constructing OBS networks. The analytical model is verified through simulation. [ABSTRACT FROM AUTHOR]

Mesh generation is one of the key issues in Computational Fluid Dynamics. This paper presents an adaptive two-dimensional mesh refinement method based on the law of mass conservation. The method can be used to a governing system that includes the law of mass conservation (continuity equation) for incompressible or compressible steady flows. We show one example that demonstrates the streamlines constructed using the refined mesh is accurate. [ABSTRACT FROM AUTHOR]

The problem of the capacity expansion of a network by means of expanding the capacity of its arcs is elaborated in this paper. By using the theory and algorithm of the maximum flow, the minimum cut-set and the bottleneck of the network can be found, then the expansible arcs are sought according to whether there exists any augmenting path in the minimum cut-set. If the expansible arcs exist,the capacity of the network can be expanded by means of expanding the capacity of those arcs to increase the maximum flow. In order to fulfill the goal, the algorithms of the expansible arc and the optimal expansion arc are presented and the complexity analyses of those algorithms are given as well. [ABSTRACT FROM AUTHOR]

Previously, we proposed a direct proportional length-based DNA computing approach for weighted graph problem. The approach has been proposed essentially to overcome the shortcoming of constant proportional length-based DNA computing approach. However, by using this approach, the minimum weight of edges that can be encoded is limited. Hence, in this paper, the lower bound, in term of minimum weight that can be encoded by direct proportional length-based DNA computing is analyzed. Also, the parameters contribute to the lower bound are investigated in order to identify the relation between those parameters and the lower bound of the direct proportional length-based DNA computing approach. [ABSTRACT FROM AUTHOR]

The primary goal of this paper is to apply the Half-Sweep Iterative Alternating Decomposition Explicit (HSIADE) method for solving one-dimensional diffusion problems. The formulation of the HSIADE method is also derived. Some numerical experiments are conducted that to verify the HSIADE method is more efficient than the Full-Sweep method. [ABSTRACT FROM AUTHOR]

This paper describes a distributed algorithm for use in the analysis of the multi-spectral satellite images. The algorithm combines the spatial-frequency wavelet transform and the maximum selection fusion algorithm. It fuses a multi-spectral image set into a single high quality achromatic image that suits the human visual system. The algorithm operates on a distributed collection of computers that are connected through LAN. A scene of Chiangmai, Thailand, taken from the Landsat ETM+ sensor is used to assess the algorithm image quality, performance and scalability. [ABSTRACT FROM AUTHOR]

In this paper, a novel WNN, multi-input and multi-output feedforward wavelet neural network is constructed. In the hidden layer, wavelet basis functions are used as activate function instead of the sigmoid function of feedforward network. The training formulas based on BP algorithm are mathematically derived and training algorithm is presented. A numerical experiment is given to validate the application of this wavelet neural network in multi-variable functional approximation. [ABSTRACT FROM AUTHOR]

In this paper, an overview of existing MAC scheduling algorithms is briefly presented. Then considerations concerning QoS of wireless communication are discussed. In order to improve the QoS of wireless communication, we proposed a new QoS-based algorithm for wireless access and scheduling in centralized network where all traffics are controlled by the base station. The algorithm uses TDMA in the radio air interface, but it can also be adapted to other access methods, such as CDMA. Finally, the simulation is performed to show the QoS guarantee with the new algorithm. [ABSTRACT FROM AUTHOR]

Let n, s be positive integers such that . A directed double-loop network G(n ; 1, s) is a directed graph (V, E), where V = Zn = {0, 1, 2, ... , n-1 }, E = { i, =0, 1, 2, ... , n-1 }. An attention deserving problem is: for a given positive integer n, how to choose an integer s such that the diameter of G(n; 1, s) is minimal. An algorithm, where O(k) is upper-bounded by O(n0.25), is given in this paper to determine s such that the diameter of G(n; 1, s) is minimal. Keywords: directed double-loop networks, diameter, optimal design, k-tight optimal. [ABSTRACT FROM AUTHOR]

Grid computing is recently developed concept in high performance computing. Security and resource management is the two key technologies in a grid. In this paper, we describe a grid portal model based on security proxy and storage resource broker. We set up a testing environment and make some improvements to the proxy model. We also point out several disadvantages of current mechanism, and make some corresponding improvements. Keywords: Grid Computing, Grid Portal, Security Proxy, Storage and Data Management. [ABSTRACT FROM AUTHOR]

A new computability methodology was proposed for the fuzzy proportional integral derivative (PID) controllers based on the theoretical fuzzy analysis and the downhill simplex optimization. The paper analyzes the algorithm of downhill simplex searching of the optimization objective functions. The input and objective function of downhill factors were selected for constructing optimal decision rules for the fuzzy logic controller. An optimizer was built around the simplex algorithm that it minimized a simplex within an N-dimensional. The sampling rate is 0.1 and the controllers are implemented under a 0.5 second time delay. The simulation confirmed the viability of the algorithm in its effectiveness of the adaptive fuzzy logic controller. [ABSTRACT FROM AUTHOR]