The Experts below are selected from a list of 327 Experts worldwide ranked by ideXlab platform
J Schijve - One of the best experts on this subject based on the ideXlab platform.
-
Statistical Distribution functions and fatigue of structures
International Journal of Fatigue, 2005Co-Authors: J SchijveAbstract:Abstract Three Statistical Distribution functions are compared. The log(N) normal Distribution, the 3-parameter Weibull Distribution and the 3-parameter log( N − N 0 ) normal Distribution. The latter function is relatively new. Attention is paid to very low probabilities of failure. Various sources of scatter of fatigue test data and fatigue of structures in service are recapitulated. Different practical problems for which statistics are important are defined. Limitations of Statistical predictions are briefly discussed. The significance of fatigue acceptance tests and realistic full scale service-simulation tests are emphasized.
K. Hirose - One of the best experts on this subject based on the ideXlab platform.
-
On the effectiveness of MFCCs and their Statistical Distribution properties in speaker identification
2004 IEEE Symposium on Virtual Environments Human-Computer Interfaces and Measurement Systems 2004. (VCIMS)., 2004Co-Authors: K.i. Molla, K. HiroseAbstract:This paper presents a study on the effectiveness of mel-frequency cepstrum coefficients (MFCCs) and some of their Statistical Distribution properties (skewness, kurtosis, standard deviation) as the features for text-dependent speaker identification. Multi-layer neural network with backpropagation learning algorithm is used here as the classification tool. The MFCCs representing the speaker characteristics of a speech segment are computed by nonlinear filterbank analysis and discrete cosine transform. The speaker identification efficiency and the convergence speed of the neural network are investigated for different combinations of the proposed features. The result shows that the first MFCC degrades the identification competence and Statistical Distribution parameters enhance the training speed of the neural network.
-
VECIMS - On the effectiveness of MFCCs and their Statistical Distribution properties in speaker identification
2004 IEEE Symposium on Virtual Environments Human-Computer Interfaces and Measurement Systems 2004. (VCIMS)., 2004Co-Authors: K.i. Molla, K. HiroseAbstract:This paper presents a study on the effectiveness of mel-frequency cepstrum coefficients (MFCCs) and some of their Statistical Distribution properties (skewness, kurtosis, standard deviation) as the features for text-dependent speaker identification. Multi-layer neural network with backpropagation learning algorithm is used here as the classification tool. The MFCCs representing the speaker characteristics of a speech segment are computed by nonlinear filterbank analysis and discrete cosine transform. The speaker identification efficiency and the convergence speed of the neural network are investigated for different combinations of the proposed features. The result shows that the first MFCC degrades the identification competence and Statistical Distribution parameters enhance the training speed of the neural network.
K.i. Molla - One of the best experts on this subject based on the ideXlab platform.
-
On the effectiveness of MFCCs and their Statistical Distribution properties in speaker identification
2004 IEEE Symposium on Virtual Environments Human-Computer Interfaces and Measurement Systems 2004. (VCIMS)., 2004Co-Authors: K.i. Molla, K. HiroseAbstract:This paper presents a study on the effectiveness of mel-frequency cepstrum coefficients (MFCCs) and some of their Statistical Distribution properties (skewness, kurtosis, standard deviation) as the features for text-dependent speaker identification. Multi-layer neural network with backpropagation learning algorithm is used here as the classification tool. The MFCCs representing the speaker characteristics of a speech segment are computed by nonlinear filterbank analysis and discrete cosine transform. The speaker identification efficiency and the convergence speed of the neural network are investigated for different combinations of the proposed features. The result shows that the first MFCC degrades the identification competence and Statistical Distribution parameters enhance the training speed of the neural network.
-
VECIMS - On the effectiveness of MFCCs and their Statistical Distribution properties in speaker identification
2004 IEEE Symposium on Virtual Environments Human-Computer Interfaces and Measurement Systems 2004. (VCIMS)., 2004Co-Authors: K.i. Molla, K. HiroseAbstract:This paper presents a study on the effectiveness of mel-frequency cepstrum coefficients (MFCCs) and some of their Statistical Distribution properties (skewness, kurtosis, standard deviation) as the features for text-dependent speaker identification. Multi-layer neural network with backpropagation learning algorithm is used here as the classification tool. The MFCCs representing the speaker characteristics of a speech segment are computed by nonlinear filterbank analysis and discrete cosine transform. The speaker identification efficiency and the convergence speed of the neural network are investigated for different combinations of the proposed features. The result shows that the first MFCC degrades the identification competence and Statistical Distribution parameters enhance the training speed of the neural network.
Sonia Idjimarene - One of the best experts on this subject based on the ideXlab platform.
-
power laws behavior in multi state elastic models with different constraints in the Statistical Distribution of elements
Communications in Nonlinear Science and Numerical Simulation, 2014Co-Authors: Marco Scalerandi, Antonio Gliozzi, Sonia IdjimareneAbstract:Abstract Often materials exhibit nonlinearity and hysteresis in their response to an elastic excitation and the dependence of the nonlinear indicator on the excitation energy is a power law function. From the theoretical point of view, such behavior could be described using multistate elastic models based on a generalized Preisach–Mayergoyz (PM) approach. In these models a Statistical Distribution of transition parameters is usually introduced. We show in this paper the existence of a link between the power law exponent predicted by the model and the properties of the chosen Distribution. Numerical results are discussed, based on an implementation in the PM formalism of an adhesion model.
Jordi Suñé - One of the best experts on this subject based on the ideXlab platform.
-
On the progressive breakdown Statistical Distribution and its voltage acceleration
2007 IEEE International Electron Devices Meeting, 2007Co-Authors: Ernest Wu, Santi Tous, Jordi SuñéAbstract:The Statistical Distribution of progressive breakdown (PBD) time is found to be lognormal at high values of failure currents in large sample size experiments. A physics-based model captures the main experimental trends. While we confirm that the voltage scaling of PBD time is nicely modelled by a power law, the value of the power-law exponent is reported to depend on the failure current and to take values somewhat smaller than those previously reported for the first breakdown voltage acceleration.
-
The Statistical Distribution of breakdown from multiple breakdown events in one sample
Journal of Physics D, 1991Co-Authors: E. Farrés, Montserrat Nafría, Jordi Suñé, Xavier AymerichAbstract:Several breakdown events have been observed in ultrathin SiO2 layers when subjected to constant-voltage stress. It has been shown that the I(V) and I(t) characteristics can be understood assuming that the sample is formed by a great number of independent capacitors (spots) connected in parallel. Each breakdown event corresponds to the dielectric failure of one of these spots. The failure rate of the spots is related to the failure rate of the samples by the well known area effect of the breakdown Statistical Distribution. It is demonstrated that the failure rate of the samples is related to the number of broken spots and that it can be directly obtained from the evolution of the current with time. So, the Statistical Distribution of the first breakdown of a set of capacitors can be obtained by provoking multiple breakdown events in only one sample during a constant-voltage stress. Experimental results are presented which demonstrate that this is a very powerful alternative technique to measure experimentally the Statistical Distribution of breakdown in wearout tests.
