• Journal of the Korean Statistical Society
• /
• v.20 no.1
• /
• pp.29-43
• /
• 1991

Two data-based bandwidth selectors which are optimal in the sense that they achieve n$\^$-$\frac{1}{2}$/ rate of convergence in kernel density estimation are proposed. The proposed bandwidth selectors are constructed by modifying Park and Marron's plug-in method. The first modification is taking Taylor expansion of the mean integrated squared error to two more terms than in the case of plug-in method. The second is estimating more accurately the functionals of the unknown density appeared in the minimizer of the expansion by using higher order kernels. The proposed bandwidth selectors were proved to be optimal in terms of convergence rate. According to small-sample Monte Carlo studies, the proposed bandwidth selectors showed better performance than all the other bandwidth selectors considered in the simulation.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.32 no.2
• /
• pp.92-99
• /
• 2008

The "energy separation phenomenon" through a vortex tube has been a long-standing mechanical engineering problem whose operational principle is not yet known. In order to find the operational principle of the vortex tube, CFD analysis of the flow field in the vortex tube has been carried out. It was found that the energy separation mechanism in the vortex tube consists of basically two major thermodynamic-fluid mechanical processes. One is the isentropic expansion process at the inlet nozzle, during which the gas temperature is nearly isentropically cooled. Second process is the viscous dissipation heating due to the high level of turbulence in both flow passages toward cold gas exit as well as the hot gas exit of the vortex tube. Since the amount of such a viscous heating is different between the two passages, the gas temperature at the cold exit is much lower than that at the hot exit.

• 한국전산유체공학회:학술대회논문집
• /
• 1996.05a
• /
• pp.68-73
• /
• 1996

The full incompressible Navier-Stokes equations are numerically integrated to solve the unsteady channel flow using a new numerical scheme of second-order accuracy developed by the authors. It is well known that in spite of the symmetry in the boundary condition and geometry, asymmetry can develop with time-dependency in a channel with sudden expansion. The instability of the shear flow and the cross-channel pressure contribute to such asymmetric flow. In this paper, we successfully generated a channel flow in which vortex waves were propagated downstream due to the harmonically oscillating inlet flow. The structure of the eddies and wall vorticity are parametrically investigated.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.327-330
• /
• 1998

Wave propagation in inhomogeneous elastic media is studied by using the Fourier method, where the spatial derivatives are computed by the FFT algorithm, while the time derivatives are expanded into the second order finite different expansion. For numerical examples, wave propagation in the layered half-plane are investigated. The comparisons of numerical and analytic results shows good agreement.

• Proceedings of the Safety Management and Science Conference
• /
• 2001.05a
• /
• pp.179-184
• /
• 2001

The objective of this work is to develop the capability to analyze accurately the mixed-mode propagation of a crack in composite structures with elastic orthotropic material stiffness properties and anisotropic material strength characteristics. We employ the normal stress ratio theory to predict the direction of crack extension. It is shown that the angle of crack extension can be altered by the use of second order term in the series expansion is important for the accurate determination of crack growth direction.

• Communications for Statistical Applications and Methods
• /
• v.7 no.1
• /
• pp.233-244
• /
• 2000

Confidence intervals for survival function give useful information about the lifetime distribution. In this paper we develop Edgeworkth expansions as approximation to the true and bootstrap distributions of normalized nonparametric maximum likelihood estimator of survival function in the Koziol-Green model and then use these results to show that the bootstrap approximations have second order accuracy.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2006.11a
• /
• pp.175-177
• /
• 2006

This letter deals with non-pilot-aided symbol timing estimation methods in an orthogonal frequency division multiplexing (OFDM) system. To do this, OFDM system uses a frequency diversity scheme over two consecutive data symbols. Our approach can be viewed as an expansion of Schmidl's and Minn's correlation methods. Using the OFDM signal equipped with frequency diversity, however, symbol timing is accurately estimated without additional training symbol and a second-order diversity gain is achieved.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2956-2958
• /
• 1999

The classical solution to the noise removal problem is the Wiener filter, which utilizes the second-order statistics of the Fourier decomposition. We discuss a Bayesian formalism which gives rise to a type of wavelet threshold estimation in non-parametric regression. A prior distribution is imposed on the wavelet coefficients of the unknown response function, designed to capture the sparseness of wavelet expansion common to most application. For the prior specified, the posterior median yields a thresholding procedure

• Bulletin of the Korean Chemical Society
• /
• v.10 no.5
• /
• pp.452-458
• /
• 1989

The static and dynamic phenomena of a model for the metabolic control system with positive feedback are discussed with the static and dynamic renormalization group theory. Then, the explicit results for the static and dynamic exponents are obtained up to the second order of ${\varepsilon}$-expansion, ${\varepsilon}$ being 4-d, where d is the space dimensionality of the system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.2
• /
• pp.483-494
• /
• 1996

In this paper, we proposed the receive decoder and Virterbi algorithm with sliding multiple symbol detection using MLSE. the informationis transmitted by the phase difference of the adjacent channel signal at the .pi./4 shift QPSK. In order to apply the .pi./4 shift QPSK to TCM, we use the signal set expansion and the signal set partition by the phase differences. And the Viterbi decoder containing branch mertrice of the squared Euclidean distance of the first, second and Lth order phase difference is introduced in order to extract the information in the differential detection of the Trellis-Coded .pi./4 shift QPSK. The proposed Viterbi decoder and receiver are conceptually same to the sliding multiple symbol detection method using the MLSE. By uisng this method, the study shows that the Trellis-Coded .pi./4 shift QPSK is an attractive scheme for the power and the bandimited systems while also improving the BER performance when the Viterbi decoder is employed to the Lth order phase difference metrics.