• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.286-288
• /
• 2005

최근 기지국(Base Station)의 도움 없이 이동 단말기 간의 다중 무선 홉을 사용하여 송,수신자 간의 데이터 전송을 가능하게 하는 Ad-hoc 통신망에 관한 연구가 활발히 진행되고 있다. 유선망과 달리 Ad-hoc 통신망은 무선 전송 매체를 사용하기 때문에 신호의 페이딩(Fading), 간섭(Interference), 잡음(Noise) 등에 의해 높은 BER(Bit Error Rate)이 발생하는 특징을 가지고 있다. 하지만, 현재 인터넷 상에서 광범위하게 사용되고 있는 전송 규약인 TCP(Transmission Control Protocol)는 유선망의 신뢰적인 전송 매체를 고려하여 개발된 프로토콜이기 때문에 TCP를 수정 없이 Ad-hoc 통신망에 적용할 경우 전송 성능이 저하되는 문제를 가지고 있다. 전송 성능이 저하되는 문제는 기존 TCP가 에러 발생의 원인을 혼잡에 의한 것으로 인식하고 불필요한 혼잡 제어를 하기 때문이다. 본 논문에서는 송신자가 에러 발생 원인을 구별하고, 그에 따라 전송률을 조절함으로써 Ad-hoc 망에서의 TCP성능 향상을 위한 방법을 제시하였다. 또한 ns-2 시뮬레이터를 이용한 실험을 통해 TCP의 성능이 제안된 알고리즘에 의해 향상되었음을 확인하였다

• The Journal of Information Technology
• /
• v.2 no.2
• /
• pp.139-144
• /
• 1999

This thesis is a study on using of entropy information about the additional words in the after processing step to promote an accuracy in speech recognition system. The exsisting ratio of Woodo detective method changes the efficiency of speech recognition system according to speech data and increases the probability of producing error recognition because of similarity of value of Woodo in the additional words. But we could obtain the accurate speech recognition system which heightens discrimination becoming independent of speech data by using of after processing method refusing a candidate which entropy price is lower among words except words we could recognize than entropy Price of each additional word. As a result of this experiment when the false alarm is 20 percent, we could put out the maximum 3.6 percent efficiency of recognition system through this after processing method by entropy more than the method by ratio of Woods.

• Journal of The Korean Association For Science Education
• /
• v.13 no.1
• /
• pp.80-91
• /
• 1993

The purpose of this study is to develop an reliable instrument for evaluating science inquiry skills through an R&D procedure. A total of ten seience inquiry skills were selected for the development of the instrument, ie, observation, classifying, recognizing problems, measuring, inferring, formulating hypothesis, controlling variables, experimenting, interpreting and drowing a conclusion. And three items were developed for each sceince inquiry skill, totaling up to thirty items. The content areas of developed items are divided into three categories, material and energy, life and environment, and the earth and the space. There are 10 items for each category. The content validity and the objectivity of developed items were checked, verified in the two field trials and revised according to the analysis of items by seven experienced specialists with the degree of doctors in science education and five teachers who were studying for their master's degree in science education. According to the results of the second field trial. the content validity of the instrument was 91.1%, reliablity(K-R 20) 0.78, defficulty index 49.13%, discrimination index 0.42, mean achivement 61.85%, standard deviation 5.11, and standard error 2.40. Considering the above results, the developed instrument in this study is regarded as a appropriate instrument for evaluating science inquiry skills of elementary school students.

• Journal of Ginseng Research
• /
• v.22 no.2
• /
• pp.96-101
• /
• 1998

This paper describes the determination of bulk density and the discrimination of internal structure of red ginseng by nuclear magnetic resonance (NMR). The 102 red ginseng roots were tested for bulk density. The NMR properties measured by NMR parameters such as spin-lattice relaxation time ($T_1$) and spin-spin relaxation time ($T_2$) were determined using the low field proton NMR analyzer. Bulk density of red ginseng root showed a highly negative significant correlation (r=-0.8934) with the value of $T_1$, but a highly positive significant correlation (r=0.7672 and 0.5909) with the value of T21 (short T2) and T22 (long T2), respectively. Multiple regression equation, Y=-0.0069.$T_1$+0.3044.$T_{21}$-0.0156.$T_{22}$-0.6368, using the MNR parameter values of 80 red ginseng roots can effectively predict the bulk density of 22 red ginseng roots with the correlation coefficient of 0.9396 and the standard error of 0.086. The differences in the internal structure of normal and inside white part of red ginseng were easily found by the signal intensity of NMR image based on magnetic properties of proton nucleus.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.1
• /
• pp.115-122
• /
• 1999

In this paper, we confirmed that a speech signal is a chaos signal, and in order to use it as a speaker recognition parameter, analyzed chaos dimension. In order to raise speaker identification and pattern recognition, by making up the strange attractor involving an individual's vocal tract characteristics very well and applying fuzzy membership function to correlation dimension, we proposed fuzzy correlation dimension. By estimating the correlation of the points making up an attractor are limited according space dimension value, fuzzy correlation dimension absorbed the variation of the reference pattern attractor and test pattern attractor. Concerning fuzzy correlation dimension, by estimating the distance according to the average value of discrimination error per each speaker and reference pattern, investigated the validity of speaker recognition parameter.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.295-300
• /
• 1998

In this paper, a real time diagnostic algorithm fur estimating the impact location by loose parts is proposed. It is composed of two modules such as the alarm discrimination module (ADM) and the impact-location estimation module(IEM). ADM decides whether the detected signal that triggers the alarm is the impact signal by loose parts or the noise signal. When the decision from ADM is concluded as the impact signal, the beginning time of burst-type signal, which the impact signal has usually such a form in time domain, provides the necessary data fur IEM. IEM by use of the arrival time method estimates the impact location of loose parts. The overall results of the estimated impact location are displayed on a computer monitor by the graphical mode and numerical data composed of the impact point, and thereby a plant operator can recognize easily the status of the impact event. This algorithm can perform the diagnosis process automatically and hence the operator's burden and the possible operator's error due to lack of expert knowledge of impact signals can be reduced remarkably. In order to validate the application of this method, the test experiment with a mock-up (flat board and reactor) system is performed. The experimental results show the efficiency of this algorithm even under high level noise and potential application to Loose Part Monitoring System (LPMS) for improving diagnosis capability in nuclear power plants.

• Phonetics and Speech Sciences
• /
• v.2 no.2
• /
• pp.63-68
• /
• 2010

A new version of sequential probability ratio test (SPRT) which has been investigated in utterance-length control is proposed to obtain uniform response results in speaker verification (SV). Although SPRTs can obtain fast responses in SV tests, differences in the performance may occur depending on the compositions of consonants and vowels in the sentences used. In this paper, a fast sequential probability ratio test (FSPRT) method that shows consistent performances at all times regardless of the compositions of vocalized sentences for SV will be proposed. In generating frames, the FSPRT will first conduct SV test processes with only generated frames without any overlapping and if the results do not satisfy discrimination criteria, the FSPRT will sequentially use frames applied with overlapping. With the progress of processes as such, the test will not be affected by the compositions of sentences for SV and thus fast response outcomes and even consistent performances can be obtained. Experimental results show that the FSPRT has better performance to the SPRT method while requiring less complexity with equal error rates (EER).

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.576-580
• /
• 2005

New real-time confocal microscopy using spectral encoding technique and slit confocal aperture is proposed and designed. Spectral encoding technique, which encodes one-dimensional spatial information of a specimen in wavelength, and slit aperture make it possible to obtain two-dimensional lateral image of the specimen simultaneously at standard video rates without expensive scanning units such as polygon mirrors and galvano mirrors. The working principle and the configuration of the system are explained. The variation in axial responses for the simplified model of the system with normalized slit width is numerically analyzed based on the wave optics theory. Slit width that directly affects the depth discrimination of the system is determined by a compromise between axial resolution and signal intensity from the simulation result. On the assumption of the lateral sampling resolution of 50 nm, design variables and governing equations of the system are derived. The system is designed to have the mapping error less than the half pixel size, to be diffraction-limited and to have the maximum illumination efficiency. The designed system has the FOV of $12.8um{\times}9.6um$, the theoretical axial FWHM of 1.1 um and the lateral magnification of-367.8.

• International Journal of Computer Science & Network Security
• /
• v.21 no.11
• /
• pp.89-96
• /
• 2021

This work is an extension of my work presented a robust and economically efficient method for the Discrimination of four Mung-Beans [1] varieties based on quantitative parameters. Due to the advancement of technology, users try to find the solutions to their daily life problems using smartphones but still for computing power and memory. Hence, there is a need to find the best classifier to classify the Mung-Beans using already suggested features in previous work with minimum memory requirements and computational power. To achieve this study's goal, we take the experiments on various supervised classifiers with simple architecture and calculations and give the robust performance on the most relevant 10 suggested features selected by Fisher Co-efficient, Probability of Error, Mutual Information, and wavelet features. After the analysis, we replace the Artificial Neural Network and Deep learning with a classifier that gives approximately the same classification results as the above classifier but is efficient in terms of resources and time complexity. This classifier is easily implemented in the smartphone environment.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.117-125
• /
• 2015

In this paper, we propose a classification algorithm based on the obtained EEG(Electroencephalogram) signal for the control of 'left' and 'right' turnings of which a driving system composed of EEG sensor, Labview, DAQ, Matlab and driving robot. The proposed algorithm uses features extracted from frequency band information obtained by DWT (Discrete Wavelet Transform) and selects features of high discrimination by using Fisher score. We, also propose the number of feature vectors for the best classification performance by using SVM(Support Vector Machine) classifier and propose a decision pending algorithm based on MLD (Maximum Likelihood Decision) to prevent malfunction due to misclassification. The selected four feature vectors for the proposed algorithm are the mean of absolute value of voltage and the standard deviation of d5(2-4Hz) and d2(16-32Hz) frequency bands of P8 channel according to the international standard electrode placement method. By using the SVM classifier, we obtained 98.75% accuracy and 1.25% error rate. Also, when we specify error probability of 70% for decision pending, we obtained 95.63% accuracy and 0% error rate by using the proposed decision pending algorithm.