• Communications for Statistical Applications and Methods
• /
• v.3 no.3
• /
• pp.215-223
• /
• 1996

By using Tukey's generalized lambda distribution, appoximate posterior density is derived under the Bayes-empirical Bayes model. The sensitivity of posterior distribution to the hyperprior distribution is examined by using Tukey's generalized lambda distriburion which approximate many well-knmown distributions. Based upon Monte Varlo simulation studies it can be said that posterior distribution is sensitive to the cariance of the prior distribution and to the symmetry of the hyperprior distribution. Also posterior distribution is approximately obtained by using the following methods : Lindley method, Laplace method and Gibbs sampler method.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.11
• /
• pp.36-42
• /
• 2010

This paper presents a hybrid approach to the construction of quasi-cyclic (QC) low-density parity-check (LDPC) codes based on parallel bundles in Euclidean geometries and circulant permutation matrices. Codes constructed by this method are shown to be regular with large girth and low density. Simulation results show that these codes perform very well with iterative decoding and achieve reasonably large coding gains over uncoded system.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.778-790
• /
• 2023

Although density peak clustering can often easily yield excellent results, there is still room for improvement when dealing with complex, high-dimensional datasets. One of the main limitations of this algorithm is its reliance on geometric distance as the sole similarity measurement. To address this limitation, we draw inspiration from the information bottleneck theory, and propose a novel density peak clustering algorithm that incorporates this theory as a similarity measure. Specifically, our algorithm utilizes the joint probability distribution between data objects and feature information, and employs the loss of mutual information as the measurement standard. This approach not only eliminates the potential for subjective error in selecting similarity method, but also enhances performance on datasets with multiple centers and high dimensionality. To evaluate the effectiveness of our algorithm, we conducted experiments using ten carefully selected datasets and compared the results with three other algorithms. The experimental results demonstrate that our information bottleneck-based density peaks clustering (IBDPC) algorithm consistently achieves high levels of accuracy, highlighting its potential as a valuable tool for data clustering tasks.

• Proceedings of the KSME Conference
• /
• 2007.05b
• /
• pp.3197-3202
• /
• 2007

An analysis program for pedestrian flow has been developed to investigate the flow patterns of passenger in railway stations. Analysis algorithms for pedestrian flow based on DEM(Discrete Element Method) are newly developed. There are lots of similarity between particle-laden two phase flow and passenger flow. The velocity component of 1st phase corresponds to the unit vector of calculation cell, each particle to passenger, volume fraction to population density and the particle velocity to the walking velocity, etc. And, the walking velocity of passenger is also represented by the function of population density. Key algorithms are developed to determine the position of passenger, population density and numbering to each passenger. By using the developed program, we compared the simulation results of the effects of the location and size of exit and elapsed time.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.4
• /
• pp.146-150
• /
• 2021

Plasma information based virtual metrology (PI-VM) that predicts wafer-to-wafer etch rate variation after wet cleaning of plasma facing parts was developed. As input parameters, plasma information (PI) variables such as electron temperature, fluorine density and hydrogen density were extracted from optical emission spectroscopy (OES) data for etch plasma. The PI-VM model was trained by stepwise variable selection method and multi-linear regression method. The expected etch rate by PI-VM showed high correlation coefficient with measured etch rate from SEM image analysis. The PI-VM model revealed that the root cause of etch rate variation after the wet cleaning was desorption of hydrogen from the cleaned parts as hydrogen combined with fluorine and decreased etchant density and etch rate.

• Journal of Mechanical Science and Technology
• /
• v.19 no.12
• /
• pp.2253-2262
• /
• 2005

The focus of this work is placed on the analysis of the mixture formation mechanism under the evaporative diesel spray of impinging and free conditions. As an experimental parameter, ambient gas density was selected. Effects of density variation of ambient gas on liquid and vapor-phase inside structure of evaporation diesel spray were investigated. Ambient gas density was changed between ${\rho}a=5.0\;kg/m^3$ and $12.3\;kg/m^3$. In the case of impinging spray, the spray spreading to the radial direction is larger due to the decrease of drag force of ambient gas in the case of the low density than that of the high density. On the other hand, in the case of free spray, in accordance with the increase in the ambient gas density, the liquid-phase length is getting short due to the increase in drag force of ambient gas. In order to examine the homogeneity of mixture consisted of vapor-phase fuel and ambient gas in the spray, image analysis was conducted with statistical thermodynamics based on the non-dimensional entropy (S) method. In the case of application of entropy analysis to diesel spray, the entropy value always increases. The entropy of higher ambient density is higher than that of lower ambient gas density during initial injection period.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.6
• /
• pp.619-625
• /
• 2009

FastSLAM which uses the Rao-Blackwellized particle filter is one of the famous solutions to SLAM (Simultaneous Localization and Mapping) problem that estimates concurrently a robot's pose and surrounding environment. However, the particle depletion problem arises from the loss of the particle diversity in the resampling process of FastSLAM. Then, the performance of FastSLAM degenerates over the time. In this work, DIR (Density Information-based Resampling) technique is proposed to solve the particle depletion problem. First, the cluster is constructed based on the density of each particle, and the density of each cluster is computed. After that, the number of particles to be reserved in each cluster is determined using a linear method based on the distance between the highest density cluster and each cluster. Finally, the resampling process is performed by rejecting the particles which are not selected to be reserved in each cluster. The performance of the DIR proposed to solve the particle depletion problem in FastSLAM was verified in computer simulations, which significantly reduced both the RMS position error and the feature error.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.05a
• /
• pp.672-676
• /
• 2002

In this paper, a reliability-based design optimization method, which enables the determination of optimum design that incorporate confidence range for structures, is studied. Response surface method and Monte Carlo simulation are utilized to determine limit state function. The proposed method is applied to the I-type steel structure for reliability based optimal design.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.27 no.2
• /
• pp.132-135
• /
• 2014

The Sr based ceramic thin films were deposited on Si substrate by RF magnetron sputtering method. And Sr based thin films were annealed at $500{\sim}700^{\circ}C$ using RTA. The surface roughness showed about 2.4 nm in annealed thin film at $600^{\circ}C$. The capacitance density of Sr based thin films were increased with the increase of annealing temperature. The maximum capacitance density of $0.6{\mu}F/cm^2$ was obtained by annealing temperature at $700^{\circ}C$. The voltage dependence of dielectric loss showed about 0.02 in voltage ranges of -10~+10 V. The leakage current density of annealing temperature of $600^{\circ}C$ was the $4.0{\times}10^{-6}\;A/cm^2$ at applied voltage of -5~+5 V.

• Journal of Biomedical Engineering Research
• /
• v.21 no.4
• /
• pp.373-384
• /
• 2000

The goal of this paper is to develop a computational method to predict cancellous bone density distributions based upon continuum levels of volumetric strain. Volumetric strain is defined as the summation of normal strains, excluding shear strains, within an elastic range of loadings. Volumetric strain at a particular location in a cancellous structure changes with changes of the boundary conditions (prescribed displacements, tractions, and pressure). This change in the volumetric strain is postulated to predict the adaptive change in the bone apparent density. This bone remodeling theory based on volumetric strain is then used with the finite element method to compute the apparent density distribution for cancellous bone in both lumbar spine and proximal femur using an iterative algorithm, considering the dead zone of strain stimuli. The apparent density distribution of cancellous bone predicted by this method has the same pattern as experimental data reported in the literature (Wolff 1892, Keller et al. 1989, Cody et al. 1992). The resulting bone apparent density distributions predict Young's modulus and strength distributions throughout cancellous bone in agreement with the literature (Keller et al. 1989, Carter and Hayes 1977). The method was convergent and sensitive to changes in boundary conditions. Therefore, the computational algorithm of the present study appears to be a useful approach to predict the apparent density distribution of cancellous bone (i.e. a numerical approximation for Wolff's Law)