• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.9
• /
• pp.62-70
• /
• 1995

In some practical applications of the LMS Algorithm the coefficient adaptation can be performed only after some fixed delay. The resulting algorithm is known as the Delayed Least Mean Square (DLMS) algorithm in the literature. There exist analyses for this algorithm, but most of them are based on the unrealistic independence assumption between successive input vectors. Inthis paper we consider the DLMS algorithm with decreasing step size .mu.(n)=n/a, a>0 and prove the almost-sure convergence ofthe weight vector W(n) to the Wiener solution W$_{opt}$ as n .rarw. .inf. under the mixing unput condition and the satisfaction of the law of large numbers. Computer simulations for decision-directed adaptive equalizer with decoding delay are performed to demonstrate the functioning of the proposed algorithm.m.

• Speech Sciences
• /
• v.1
• /
• pp.277-283
• /
• 1997

In this paper, we present a robust recognition algorithm based on the Wiener filtering method as a research tool to develop the Korean Speech recognition system. We especially used Wiener filtering method in cepstrum-domain, because the method in frequency-domain is computationally expensive and complex. Evaluation of the effectiveness of this method has been conducted in speaker-independent isolated Korean digit recognition tasks using discrete HMM speech recognition systems. In these tasks, we used 12th order weighted cepstral as a feature vector and added computer simulated white gaussian noise of different levels to clean speech signals for recognition experiments under noisy conditions. Experimental results show that the presented algorithm can provide an improvement in recognition of as much as from $5\%\;to\;\20\%$ in comparison to spectral subtraction method.

• International Journal of CAD/CAM
• /
• v.1 no.1
• /
• pp.23-32
• /
• 2002

In this paper, a new technique of space deformation for parametric surfaces with so-called extension function (EF) is presented. Firstly, a special extension function is introduced. Then an operator matrix is constructed on the basis of EF. Finally the deformation of a surface is achieved through multiplying the equation of the surface by an operator matrix or adding the multiplication of some vector and the operator matrix to the equation. Interactively modifying control parameters, ideal deformation effect can be got. The implementation shows that the method is simple, intuitive and easy to control. It can be used in such fields as geometric modeling and computer animation.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.4
• /
• pp.825-832
• /
• 2012

We propose a novel way of forecasting the market shares of several brands simultaneously in a multiplicative competitive interaction model, which uses kernel regression technique incorporated with kernel machine technique applied in support vector machines and other machine learning techniques. Traditionally, the estimations of the market share attraction model are performed via a maximum likelihood estimation procedure under the assumption that the data are drawn from a normal distribution. The proposed method is shown to be a good candidate for forecasting method of the market share attraction model when normal distribution is not assumed. We apply the proposed method to forecast the market shares of 4 Korean car brands simultaneously and represent better performances than maximum likelihood estimation procedure.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.5
• /
• pp.864-872
• /
• 1992

Two double-exposure holograms are made in the different view angle at the same time, using laser, by overlapping before and after the static deformation. These images are transfered to the computer. The fringe patterns of holograms are recognized by image processing and each component of the displacement and strain at each point of the specimen is obtained by vector analysis, and the results were presented in the graphical form. For the verification of all the ment process, the two experimental cases, the in-plane displacement by tension load and the out-of-plane displacement by bending load, are measured. These results are compared with the exact solution.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.1133-1136
• /
• 2005

The MPEG/audio standard results from three years of co-work by an international committe of high-fidelity audio compression experts in the Moving Picture Experts Group (MPEG/audio). The MPEG standard is rigid only where necessary to ensure interoperability. In this paper, a new approach of training and classification of acoustic signal is addressed. This is some what a fields of application aspects rather than technonical problems such as MPEG/codec, MIDI. In preprocessing, acoustic signal is transformmed using DWT so as to extract a feature parameters of sound such as loudness, pitch, bandwidth and harmonicity. these accoustic parameters are exploited to the input vector of neural perceptron. Experimental results showed that proposed approach can be used for tunning the dissonance chord.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.7
• /
• pp.437-442
• /
• 2005

In order to realize the speed control of an induction motor, the information of the rotor speed is needed. So the speed sensor as an encoder or a pulse generator is used to obtain it. But the use of speed sensor occur the some problems in the control system of an induction motor. To solve the problems, the appropriate speed estimation algorithm is used instead of the speed sensor. Also there is the limitation to improve the speed control performance of an induction motor using the existing speed estimation algorithm. Therefore, in this paper, intelligent speed estimator using Fuzzy-Neural systems as adaptive laws in Model Reference Adaptive System is proposed so as to improve the existing estimation algorithm and ,using the rotor speed estimated by the Proposed estimator, the speed control of an induction motor without speed sensor is performed. The computer simulation and the experiment is executed to prove the performance of the speed control system usinu the proposed speed estimator.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.48 no.8
• /
• pp.426-430
• /
• 1999

In this paper, a speed sensorless algorithm for a high-speed induction motor is proposed. The proposed algorithm simply estimates rotor speed by integrating the deviation between the command current value of a controller and the real current value of the motor. To estimate rotor speed without a speed sensor, a fuzzy speed controller and a neural network speed estimator are applied. Computer simulation and implementation of the proposed system is described.

• Journal of Power Electronics
• /
• v.15 no.1
• /
• pp.116-122
• /
• 2015

This paper presents a new current sharing control strategy for parallel-connected, synchronised three-phase DC-AC converters employing space vector pulse width modulation (SVPWM) without current sharing reactors. Unlike conventional control methods, the proposed method breaks the paths of the circulating current by dividing the switching cycle evenly between parallel connected equally rated converters. Accordingly, any inter-module reactors or circulating current control will be redundant, leading to reductions in system costs, size, and control algorithm complexity. Each converter in the new scheme employs a synchronous dq current regulator that uses only local information to attain a desired converter current. A stability analysis of the current controller is included together with a simulation of the converter and load current waveforms. Experimental results from a 2.5kVA test rig are included to verify the proposed control method.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.10
• /
• pp.3155-3163
• /
• 2000

In the recognitionsystem of car license plate, it is very important to extract the character from the license plate and recognize the extrated character. In this paper, I use the adaptive algorithm to recognize the charactor of licensse plate image. The adaptive algorithm is compounded of thinning algorithm template matching,algarthm, vector algorithm and so on. The adaptive algorithm was used to recognize the character from license image. In the result of expenment, character recognition is about up to 90% with the adaptive algorithm for the character region.