• ETRI Journal
• /
• v.32 no.5
• /
• pp.795-800
• /
• 2010

Embedding a large vocabulary speech recognition system in mobile devices requires a reduced acoustic model obtained by eliminating redundant model parameters. In conventional optimization methods based on the minimum description length (MDL) criterion, a binary Gaussian tree is built at each state of a hidden Markov model by iteratively finding and merging similar mixture components. An optimal subset of the tree nodes is then selected to generate a downsized acoustic model. To obtain a better binary Gaussian tree by improving the process of finding the most similar Gaussian components, this paper proposes a new distance measure that exploits the difference in likelihood values for cases before and after two components are combined. The mixture weight of Gaussian components is also introduced in the component merging step. Experimental results show that the proposed method outperforms MDL-based optimization using either a Kullback-Leibler (KL) divergence or weighted KL divergence measure. The proposed method could also reduce the acoustic model size by 50% with less than a 1.5% increase in error rate compared to a baseline system.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.591_592
• /
• 2009

For a similar price signal to local consumer, a regional clustering is important that can transfer a equal sign to the bus which has each different price on the basis of bus. If you take LMP and local information this method into consideration at the same time, and you give weight on neither one side, buses on a borderline can move on arbitrary. For solving this problem, in this paper will propose a clustering method to have applied new concept called 'electric distance'.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.8
• /
• pp.1128-1135
• /
• 1993

In this paper, new binary cyclic codes (hereafter, refered to as GMW code) which are generated by using GMW sequence, g(t) = tr((trk(a))r), and its cyclic shifts are introduced. Code length of GMW codes is 2a-1, where k is composite integer, e·J. Dimension of the GMW codes is k(k/j)w-1, where w is a Hamming weight of r. Several properties of GMW codes such as designed distance, minimum distance, and weights of code words are obtained in terms of parameters of GMW sequences. And expansion of GMW sequences in terms of m-sequence and its decimation sequences are introduced and characteristic polynomials of GMW sequences are also derived.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.149-167
• /
• 2005

The minimum disparity estimators introduced by Lindsay and Basu (1994) are studied empirically. An extensive simulation in this paper provides a location estimate of the small sample and supplies empirical evidence of the estimator performance for the univariate contaminated normal model. Empirical results show that the minimum generalized negative exponential disparity estimator (MGNEDE) obtains high efficiency for small sample sizes and dominates the maximum likelihood estimator (MLE) and the minimum blended weight Hellinger distance estimator (MBWHDE) with respect to efficiency at the contaminated model.

• Proceedings of the KSR Conference
• /
• 2001.10a
• /
• pp.125-130
• /
• 2001

It is vary importance that stopping distance for the freight train and need to design parameter of the brake force and friction coefficient. Acoording to the brake force between shoe and wheel less than the adhesion between wheel and rail. Because of vary difference between empty and weight car of 0∼500kN, this solution was made to application for variable loaded brake system. When the V=110km/h. the emergency stopping distance of freight train is about 700m, so this study was considered on the two condition, one of the increse brake force and to be different of the increse friction coefficient on the brake shoe. It was useful increse friction coefficient. Result of study, analyze effect of the brake shoe on the brake force of the freight car and high friction coefficient were proposed. To do this, ${\mu}$=0.155${\pm}$10% when S=600m on the V=110km/h of the train, 2 groove of friction surface on the temperature distribution were considered.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.7
• /
• pp.101-108
• /
• 1998

In this paper, after we got to realized FDNN (fuzzy decision neural network) applied the quantization triangularity fuzzy function to DBNN(decision based neural network) of a hierarchical structure for image process, we could esign hardware of the realized FDNN. Also it is normalized the standard image and the input image as the same size. We are applied DWW algorithm which selected the closest value with finding similarity of an interval image by this distance to FDNN. So we could calulated in terms of distance to weight of pixel which composed two image and eliminated the nise of image, minimized the lost of information, obtained the optimal information. It is designed hardware of high speed weighted FDNN using COMPASS tool. Aslo, the total circuit is realized as gates of 61,000 and could show to superiority of FDNN using the simulation.

• Journal of Sensor Science and Technology
• /
• v.23 no.3
• /
• pp.158-164
• /
• 2014

In indoor sensor networks equalization algorithms based on the minimization of Euclidean distance (MED) for the distributions of constant modulus error (CME) have yielded superior performance in compensating for signal distortions induced from optical fiber links, wireless-links and for impulsive noise problems. One main drawback of MED-CME algorithms is a heavy computational burden hindering its implementation. In this paper, a recursive gradient estimation for weight updates of the MED-CME algorithm is proposed for reducing the operations $O(N^2)$ of the conventional MED-CME to O(N) at each iteration time for N data-block size. From the simulation results of the proposed recursive method producing exactly the same results as the conventional method, the proposed estimation method can be considered to be a reliable candidate for implementation of efficient receivers in indoor sensor networks.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.9
• /
• pp.8-16
• /
• 1992

This paper presents a method which can recognized the Handwritten Korean characters by using a Context-Free Grammar. The input characters are thinned in order to dwindle the mount of data, the thinned characters are converted into one-dimension strings according to six-forms. when the point of contact among phonemes is found, two phonemes are seperated respectively by marking the index mark (\) at the points. The Context-Free Grammar to input characters is classified into group grammars concerning the similarity of phonemes, input characters are parsed by making use of the Pushdown automata method. As the bent parts in the Handwritten characters are found frequently, We try to correct the bent parts by using the parsing distance measure, which recognize characters according to minium value caused by measuring the weight distance between two sentences. In this experiment, the recognition rate shows 93.8% to 275 Handwritten Korean characters.

• Journal of the East Asian Society of Dietary Life
• /
• v.18 no.3
• /
• pp.405-414
• /
• 2008

The purpose of this study was to understand the relationship between a chef's physical pain complaints and occupational circumstances. Cause-and-effect analysis of employees whose work was related to cooking was analyzed in relation to, - factor of cooking circumstances (kitchen condition, cooking operation, cooking allocation and distance of movement, cooking facilities) in order to ascertain possible links to physical pain. Cooking operations appeared to causes pain, especially in neck, shoulders, and legs. Cooking allocation and distance of movement, affected backache, eye-tiredness, and change of body-weight. Cooking space and facilities was related to, pain in waist, neck, shoulders, legs, arms, wrists, and back. Because cooking operations require extended concentrations and intensive effort, inappropriate occupational circumstances can lead to employees who usually suffer from accumulated.

• Journal of Power Electronics
• /
• v.12 no.6
• /
• pp.992-1002
• /
• 2012

In this paper, the series chopper based power train for electric vehicle is proposed for aiming the increase of one battery charge driving distance. Both the motor test bench based experiments and the chassis dynamo based experiments are tested and then analyzed for EV driving performance that the proposed power train with SAZZ chopper and Quasi-PAM control scheme is compared with or without the series chopper. It is confirmed through two kinds of experimental analyses that the series chopper power train for EV can be useful to extend the driving distance if the high efficiency and the low weight of chopper are satisfied.