• 한국통신학회논문지
• /
• 제23권8호
• /
• pp.1984-1989
• /
• 1998

Convergence properties of the stochastic gradient adaptive algorithm based on the least mean absolute third (LMAT) error criterion is presented.In particular, the performnce of the algorithmis examined and compared with least mena square (LMS) algorithm for several different probability densities of the measurement noisein a system identification mode. It is observedthat the LMAT algorithm outperforms the LMS algorithm for most of the noise probability densities, except for the case of the exponentially distributed noise.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2001년도 추계학술대회
• /
• pp.17-24
• /
• 2001

A method is proposed for efficiently estimating the mean of a population which has a linear trend. The proposed estimator is based on the centered modified systematic sampling method and the concept or interpolation. Using the expected mean square error criterion, it is shown that the proposed method is more efficient than conventional methods in most real cases.

• 한국통신학회논문지
• /
• 제22권3호
• /
• pp.423-431
• /
• 1997

A new decision tree generator MEC is proposed in this paper, which uses the difference of multi-base entropy as a consistent criterion for discretization and selection of attributes. To evaluate the performance of the proposed generator, it is compared to other generators which use criteria based on entropy and adopt different discretization styles. As an experimental result, it is shown that the proposed generator produces the most efficient classifiers, which have the least number of leaves at the same error rate, regardless of whether attribute values constituting the training set are discrete or continuous.

• 한국에너지공학회:학술대회논문집
• /
• 한국에너지공학회 1994년도 추계학술발표회 초록집
• /
• pp.146-149
• /
• 1994

A nonlinear dynamic optimization has been performed for reactor power control system of CANDU 6 nuclear power plant considering xenon, fuel and moderator temperature feedback effects. Integral-of-Time-multiplied Absolute-Error (ITAE) criterion has been used as a performance index of the system behavior. Optimum controller gain are found by searching algorithm of Sequential Quadratic Programming (SQP). System models are referenced from most recent literatures. Signal flow network construction and optimization have been done by using commercial computer software package.

• Structural Engineering and Mechanics
• /
• 제6권3호
• /
• pp.339-345
• /
• 1998

The computation size and accuracy in the boundary element method are mutually coupled and strongly influenced by the formulations in boundary discretization and integration. This aspect is studied numerically for two-dimensional elastodynamic problems in the frequency-domain. The localized nature of error is observed in the computed results. A boundary discretization criterion is examined. The number of integration points in the boundary integration is studied to find the optimum number for accuracy. Useful information is obtained concerning the optimization in boundary discretization and integration.

• 제어로봇시스템학회논문지
• /
• 제12권3호
• /
• pp.294-299
• /
• 2006

This paper studies a visual feedback control scheme for robot manipulators with camera-in-hand configurations. We design a robust controller that compensates for bounded parametric uncertainties of robot mechanical dynamics. In order to reduce steady state tracking error of the robot arms due to uncertain dynamics, integral action is included in the control input. Using the Lyapunov stability criterion, the uniform ultimate boundedness of the tracking error is proved. Simulation and experimental results with a 2-1ink robot manipulator illustrate the robustness and effectiveness of the proposed control algorithm.

• 대한전기학회논문지
• /
• 제40권3호
• /
• pp.243-249
• /
• 1991

Adaptive mesh refinement scheme is incorporated with the boundary element analysis in order to get accurate solution with relatively fewer unnowns for magnetostatic field analysis. A new andsimple posteriori local error estimate is also presented. The local error is defined as an integraktion over the element of the difference between solutions from quadratic interpolation functions and linear interpolation functions and is used as the criterion for mesh refinement. Case study with a singular point reveals that adaptive meshes are more efficient in accuracy of solutions than uniform meshs generated by dividing al the elements evenly. The adaptive meshes give much better rate of convergence in global errors than the uniform meshes.

• The Journal of the Acoustical Society of Korea
• /
• 제15권4E호
• /
• pp.21-27
• /
• 1996

In this paper, we propose a discriminative training algorithm for the stochastic segment model (SSM) in continuous speech recognition. As the SSM is usually trained by maximum likelihood estimation (MLE), a discriminative training algorithm is required to improve the recognition performance. Since the SSM does not assume the conditional independence of observation sequence as is done in hidden Markov models (HMMs), the search space for decoding an unknown input utterance is increased considerably. To reduce the computational complexity and starch space amount in an iterative training algorithm for discriminative SSMs, a hybrid architecture of SSMs and HMMs is programming using HMMs. Given the segment boundaries, the parameters of the SSM are discriminatively trained by the minimum error classification criterion based on a generalized probabilistic descent (GPD) method. With the discriminative training of the SSM, the word error rate is reduced by 17% compared with the MLE-trained SSM in speaker-independent continuous speech recognition.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제54권2호
• /
• pp.55-62
• /
• 2005

The optical CT was developed to use twisted fibers for 170 kV GIS. The fiber sensor optimized on the optical CT was wound 3 turns and twisted 4 times per a turn at the pipe with a diameter of 130 mm. The measurement error for the linearity which was satisfied the criterion of 'IEC Class-1.0' was less than ± 0.44 % at the transmission type and ± 0.82 % at the reflection type to 2,000 A. At a low current to 200 A, the measurement error was ± 0.69% at the transmission type and ± 1.38 % at the reflection type.

• Fibers and Polymers
• /
• 제6권4호
• /
• pp.348-354
• /
• 2005

These days consumers' various demands are accelerating research on apparel manufacturing system including automatic measurement, pattern generation, and clothing simulation. Accordingly, methods of reconstructing human body from point-clouds measured using a three dimensional scanning device are required for apparel CAD system to support these functions. In particular, we present in this study a human body reconstruction method focused on two issues, which are the decision of the number of control point for each sectional curve with error bound and the local knot removal for reducing the unusual concentration of control points. The approximation of sectional curves with error bounds as an approximation criterion leads all sectional curves to their own particular shapes apart from the number of control points. In addition, the application of the local knot removal to construction of human body sectional curves reduces the unusual concentration of control points effectively. The results may be used to produce an apparel CAD system as an automatic pattern generation system and a clothing simulation system through the low level control of NUBS or NURBS.