Search results for " algorithm"
showing 10 items of 2538 documents
Complexity of operations on cofinite languages
2010
International audience; We study the worst case complexity of regular operation on cofinite languages (i.e., languages whose complement is finite) and provide algorithms to compute efficiently the resulting minimal automata.
A New Genetic Approach for the Partitioning Problem in Distributed Virtual Environment Systems
2004
The Partitioning problem is a key issue in the design of Distributed Virtual Environment (DVE) systems based on a server-network architecture. This problem consist of efficiently assigning the clients of the simulation (avatars) to the system servers. Despite the existing literature proposes different evolutive approaches for solving this NP-hard problem, an approach based on genetic algorithms is considered as the current best partitioning mechanism.
Performance Evaluation of CANIT Algorithm in Presence of Congestion Losses
2001
In this paper, we analize by queuing-simulation CANIT (Congestion Avoidance with Normalized Interval of Time) algorithm performances in presence of congestion losses. In a former work [3], we proposed the algorithm (CANIT) for TCP (Transmission Control Protocol) congestion avoidance phase in order to improve fairness during this phase, and we showed that using CANIT algorithm in an environment without loss, instead of standard congestion avoidance algorithm improves both congestion avoidance fairness and bandwith utilization for long RTT connections. In this paper, we consider congestion losses and show that the fairness as well as the bandwith utilization are more efficient when using CANI…
A “Noise Gene” for Econets
1993
Genetically controlled noise is applied to the weights of neural networks trained with a genetic algorithm. Networks simulate simple organisms living in an environment Reproduction is based on the ability of each network, during its life, to respond to sensory information from the environment with appropriate motor action. Each network has an amount of noise which is genetically inherited (in the ‘noise gene’) with mutations and it varies interindividually. Noise modifies the value of a weight differently for each spreading of the activation through the network. Such noise has a positive effect on the evolutionary increase in fitness and it makes fitness less dependent on the initial choice…
An adaptive prudent-daring evolutionary algorithm for noise handling in on-line PMSM drive design
2007
This paper studies the problem of the optimal control design of permanent magnet synchronous motor (PMSM) drives taking into account the noise due to sensors and measurement devices. The problem is analyzed by means of an experimental approach which considers noisy data returned by the real plant (on-line). In other words, each fitness evaluation does not come from a computer but from a real laboratory experiment. In order to perform the optimization notwithstanding presence of the noise, this paper proposes an Adaptive Prudent- Daring Evolutionary Algorithm (APDEA). The APDEA is an evolutionary algorithm with a dynamic parameter setting. Furthermore, the APDEA employs a dynamic penalty ter…
Robust adaptive algorithm with low computational cost
2006
An adaptive algorithm, which is robust to impulsive noise, is proposed. The cost function underlying this algorithm contains a parameter that controls the immunity to impulsive noise and can be easily adapted. Moreover, weight updating involves a nonlinear function, which recently has been shown to have an efficient hardware implementation. The proposed adaptive algorithm has been successfully tested in terms of accuracy and convergence on a system-identification simulation.
Quantification of melanin and hemoglobin in humain skin from multispectral image acquisition: use of a neuronal network combined to a non-negative ma…
2012
International audience; This article presents a multispectral imaging system which, coupled with a neural network-based algorithm, reconstructs reflectance cubes. The reflectance spectra are obtained using artificial neural-netwok reconstruction which generates reflectance cubes from acquired multispectral images. Then, a blind source separation algorithm based on Non-negative Matrix Factorization is used for the decomposition of human skin absorption spectra in its main pigments: melanin and hemoglobin. The analysis is performed on reflectance spectra. The implemented source separation algorithm is based on a multiplicative coefficient upload. The goal is to represent a given spectrum as t…
Discrete wavelet transform implementation in Fourier domain for multidimensional signal
2002
Wavelet transforms are often calculated by using the Mallat algorithm. In this algorithm, a signal is decomposed by a cascade of filtering and downsampling operations. Computing time can be important but the filtering operations can be speeded up by using fast Fourier transform (FFT)-based convolutions. Since it is necessary to work in the Fourier domain when large filters are used, we present some results of Fourier-based optimization of the sampling operations. Acceleration can be obtained by expressing the samplings in the Fourier domain. The general equations of the down- and upsampling of digital multidimensional signals are given. It is shown that for special cases such as the separab…
Real-Time Vector Automata
2013
We study the computational power of real-time finite automata that have been augmented with a vector of dimension k, and programmed to multiply this vector at each step by an appropriately selected k×k matrix. Only one entry of the vector can be tested for equality to 1 at any time. Classes of languages recognized by deterministic, nondeterministic, and "blind" versions of these machines are studied and compared with each other, and the associated classes for multicounter automata, automata with multiplication, and generalized finite automata.
On Finite Satisfiability of Two-Variable First-Order Logic with Equivalence Relations
2009
We show that every finitely satisfiable two-variable first-order formula with two equivalence relations has a model of size at most triply exponential with respect to its length. Thus the finite satisfiability problem for two-variable logic over the class of structures with two equivalence relations is decidable in nondeterministic triply exponential time. We also show that replacing one of the equivalence relations in the considered class of structures by a relation which is only required to be transitive leads to undecidability. This sharpens the earlier result that two-variable logic is undecidable over the class of structures with two transitive relations.