• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.128-132
• /
• 1996

In [1], it is shown that a subclass of ETBFs, which are self-dual ETBFs, can be expressed as a weighted average of median subfiltered outputs. In this paper, we extend this result to general ETBFs. In particular, we show that any ETBF can be represented as a weighted average of minimum (or maximum) subfiltered outputs. These representations naturally lead to a subclass of ETBF, called the K-th order ETBF (K-ETBF) that employs only those subfilters whose window sizes are less than or equal to K. By designing K-ETBFs under the mean square error criterion for various values of K and applying them to restore noisy signals, the tradeoff between the performance and the complexity of this class of filters is examined.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.7
• /
• pp.1007-1014
• /
• 2003

This work deals with a controlled auto-ignition (CAI) single cylinder gasoline engine, focusing on the extension of operating conditions. The fuel is injected indirectly into electrically heated inlet air flow. In order to keep a homogeneous air-fuel mixing, the fuel injector is water-cooled by a specially designed coolant passage. Investigated are the engine performance and emission characteristics under the wide range of operating conditions such as 32 to 63 in the air-fuel ratio, 1000 to 1800 rpm in the engine speed, and 150 to 18$0^{\circ}C$ in the inlet air temperature. The compression ignition gasoline engine can be achieved that the ultra lean-burn with self-ignition of gasoline fuel by heating inlet air. For example. the allowable lean limit of air-fuel ratio is extended until 63 at engine speed of 1000 rpm and inlet air temperature of 17$0^{\circ}C$. It can be achieved that the emission concentrations of carbon monoxide, hydrocarbons and nitrogen oxide had been significantly reduced by CAI combustion compared with conventional spark ignition engine.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.22 no.3
• /
• pp.407-416
• /
• 1998

The purpose of this study was to examine the effects of brand knowledge on evaluations of brand extensions in fashion market. Here, consumer knowledge toward the parent brand was based on the brand and on the company which introduced the brand. The brand extensions were classified into brand-name extension and corporate-name extension. For this study, questionnaires were administered to 700 single women in twenties. The questionnaires were designed to measure brand extension evaluations and brand knowledge in terms of familiarity, use experience and self-assessed knowledge, evaluations of the attributes and attitudes based on the brand and corporate. Employing a sample of 621 women, data were analyzed by t-test. Major findings of this study are summarized as follows; 1) The higher the level of brand knowledge such as brand familiarity, brand use experience and self-assessed brand knoil- edge was, the higher positive effects were on the evaluations of brand-name extension. Also, evaluations of brand attributes and brand attitude positively influenced the evaluations of brand-name extension. 2) The higher the level of corporate knowledge such as corporate familiarity and use experience of product manufactured by the company was, the higher positive effects were on the evaluations of corporate-name extension. Also, evaluations of corporate attributes and attitude on corporate positively influenced the evaluations of corporate-name extension. These results demonstrate that positive knowledges and affects on the parent brand are transferred to its extended product through categorization process.

• Journal of Korean Public Health Nursing
• /
• v.14 no.2
• /
• pp.203-215
• /
• 2000

The purpose of this study is to identify the effects of the supportive program for arthritis patients. who are attending a rheumatoid arthritis clinic. To achieve this purpose. this study adopted a quasi-experimental. pre- and post-test research design. comparing experimental group with control one. Supportive program was composed of in-depth. 3 times of direct interview and 2 times of advice using telephone for 8 weeks. The object of these interviews and phone was focused on the improvement of patients' preception for health. During this period. the level of pain, hemato-immunologic indices(ESR, CRP). self-efficacy, depression. and perception for health were measured in both grooups. Data were analysed by $x^2-test$, t-test. repeated measures ANOVA and Pearson's correlations. The results were as follows : 1. There were no significant differences in physiological data. 2. The feeling of self-efficacy was significantly increased in experimental group(P=.012), 3. There was no significant differences in depression. 4. The perception for health status was significantly increased in experimental group(P=.002). Thus, the supportive program. which is focused on the close interpersonal communication. proved to be effective. This result justifies the following suggestion that the role of the nursing professionals in out-patient clinic should be extended for more qualified care for the patients.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.6C
• /
• pp.490-496
• /
• 2005

In this paper, we propose a self-constructed clustering algorithm based on inference information of the fuzzy model. This method makes it possible to automatically detect and optimize the number of cluster and parameters by using input-output data. The propose method improves the performance of clustering by extended supervised learning technique. This technique uses the output information as well as input characteristics. For effect the similarity measure in clustering, we use the TSK fuzzy model to sent the information of output. In the conceptually, we design a learning method that use to feedback the information of output to the clustering since proposed algorithm perform to separate each classes in input data space. We show effectiveness of proposed method using simulation than previous ones

• Journal of Communications and Networks
• /
• v.6 no.1
• /
• pp.9-18
• /
• 2004

Self-encoded multiple access (SEMA) is a unique realization of random spread spectrum. As the term implies, the spreading code is obtained from the random digital information source instead of the traditional pseudo noise (PN) code generators. The time-varying random codes can provide additional security in wireless communications. Multi-rate transmissions or multi-level grade of services are also easily implementable in SEMA. In this paper, we analyze the performance of SEMA in additive white Gaussian noise (AWGN) channels and Rayleigh fading channels. Differential encoding eliminates the BER effect of error propagations due to receiver detection errors. The performance of SEMA approaches the random spread spectrum discussed in literature at high signal to noise ratios. For performance improvement, we employ multiuser detection and Turbo coding. We consider a downlink synchronous system such as base station to mobile communication though the analysis can be extended to uplink communications.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1494-1507
• /
• 2018

Intelligent human identification using face information has been the research hotspot ranging from Internet of Things (IoT) application, intelligent self-service bank, intelligent surveillance to public safety and intelligent access control. Since 2D face images are usually captured from a long distance in an unconstrained environment, to fully exploit this advantage and make human recognition appropriate for wider intelligent applications with higher security and convenience, the key difficulties here include gray scale change caused by illumination variance, occlusion caused by glasses, hair or scarf, self-occlusion and deformation caused by pose or expression variation. To conquer these, many solutions have been proposed. However, most of them only improve recognition performance under one influence factor, which still cannot meet the real face recognition scenario. In this paper we propose a multi-scale parallel convolutional neural network architecture to extract deep robust facial features with high discriminative ability. Abundant experiments are conducted on CMU-PIE, extended FERET and AR database. And the experiment results show that the proposed algorithm exhibits excellent discriminative ability compared with other existing algorithms.

• The Journal of the Korea Contents Association
• /
• v.19 no.9
• /
• pp.1-10
• /
• 2019

The purpose of this study is to investigate the influence of exogenous variables on the intention to use AI speaker. An online survey was administrated to 305 AI speaker users in order to examine the effect of the personal characteristics (self-efficacy, innovativeness, suitability, and enjoyment) and social impact (social conformity and social image) on perceived usefulness and easiness. The results indicate that (1) self-efficacy and social conformity have positively effect on perceived easiness; (2) suitability and social image have positively effect on perceived usefulness whereas innovativeness has negatively effect on perceived usefulness; (3) perceived usefulness and perceived easiness have significant effect on the intention to use AI speaker.

• International Journal of Advanced Culture Technology
• /
• v.10 no.2
• /
• pp.62-68
• /
• 2022

Evaluation as learning is important for the learner competency test, and the applicable method is studied. Assessment is the role of diagnosing the current learner's status and facilitating learning through appropriate feedback. The system is insufficient to enable process-oriented evaluation in small educational institute. Focusing on becoming familiar with the AI through experience can end up simply learning how to use the tools or just playing with them rather than achieving ultimate goals of AI education. In a previous study, the experience way of AI education with PLAY model was proposed, but the assessment stage is insufficient. In this paper, we propose ELAS (Experiential K-12 AI education Learning Assessment System) for small educational institute. In order to apply the Assessment factor in in this system, the AI-factor is selected by researching the goals of the current SW education and AI education. The proposed system consists of 4 modules as Assessment-factor agent, Self-assessment agent, Question-bank agent and Assessment -analysis agent. Self-assessment learning is a powerful mechanism for improving learning for students. ELAS is extended with the experiential way of AI education model of previous study, and the teacher designs the assessment through the ELAS system. ELAS enables teachers of small institutes to automate analysis and manage data accumulation following their learning purpose. With this, it is possible to adjust the learning difficulty in curriculum design to make better for your purpose.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.2
• /
• pp.147-160
• /
• 2023

Unlike other currencies, cryptocurrency is not a currency used for general transactions, but is currently applied to various investment assets and its scope is expanding. The purpose of this study is to the effect of personal characteristics on investment intention. As a theoretical background, it was verified by applying the Extended Technical Acceptance Model (ETAM). self-confidence propensity, bandwagon propensity, risk tolerance propensity, network externality, attitude, and Investment intention were composed of variables. The research method collected data from 871 people who had experience in cryptocurrency investment through a survey and analyzed it after excluding the data of 71 people who were judged to be inappropriate. The structural equation modeling method using AMOS was used. As a result of this paper, five hypotheses were accepted as statistically significant. This study concluded that self-confidence propensity, bandwagon propensity, risk tolerance propensity, network externality, and attitude had statistically significant effects on Investment intention. In this respect, this study will be able to provide useful information for cryptocurrency research.