• Journal of KIISE:Software and Applications
• /
• v.26 no.10
• /
• pp.1143-1155
• /
• 1999

셀룰러 방식의 이동통신 시스템에서 전파의 유효신호 도달범위를 예측하기 위해서는 전파전파 모델을 이용한 예측기법이 주로 사용된다. 그러나, 전파과정에서 주변 지형지물에 의해 발생하는 전파손실은 매우 복잡한 비선형적인 특성을 가지며 수식으로는 정확한 표현이 불가능하다. 본 논문에서는 신경회로망의 함수 근사화 능력을 이용하여 전파손실 예측모델을 생성하는 방법을 제안한다. 즉, 전파손실을 송수신 안테나간의 거리, 송신안테나의 특성, 장애물 투과영향, 회절특성, 도로, 수면에 의한 영향 등과 같은 전파환경 변수들의 함수로 가정하고, 신경회로망 학습을 통하여 함수를 근사화한다. 전파환경 변수들이 신경회로망 입력으로 사용되기 위해서는 3차원 지형도와 벡터지도를 이용하여 전파의 반사, 회절, 산란 등의 물리적인 특성이 고려된 특징 추출을 통해 정량적인 수치들을 계산한다. 이와 같이 얻어진 훈련데이타를 이용한 신경회로망 학습을 통해 전파손실 모델을 완성한다. 이 모델을 이용하여 서울 도심 지역의 실제 서비스 환경에 대한 타 모델과의 비교실험결과를 통해 제안하는 모델의 우수성을 보인다.Abstract In cellular mobile communication systems, wave propagation models are used in most cases to predict cell coverage. The amount of propagation loss induced by the obstacles in the propagation path, however, is a highly non-linear function, which cannot be easily represented mathematically. In this paper, we introduce the method of producing propagation loss prediction models by function approximation using neural networks. In this method, we assume the propagation loss is a function of the relevant parameters such as the distance from the base station antenna, the specification of the transmitter antenna, obstacle profile, diffraction effect, road, and water effect. The values of these parameters are produced from the field measurement data, 3D digital terrain maps, and vector maps as its inputs by a feature extraction process, which takes into account the physical characteristics of electromagnetic waves such as reflection, diffraction and scattering. The values produced are used as the input to the neural network, which are then trained to become the propagation loss prediction model. In the experimental study, we obtain a considerable amount of improvement over COST-231 model in the prediction accuracy using this model.

• Journal of KIISE:Databases
• /
• v.30 no.2
• /
• pp.176-183
• /
• 2003

Various techniques including histograms, sampling and parametric techniques have been proposed to estimate query result sizes for the query optimization. Histogram-based techniques are the most widely used form for the selectivity estimation in relational database systems. However, in the spatio-temporal databases for the moving objects, the continual changes of the data distribution suffer the direct utilization of the state of the art histogram techniques. Specifically for the future queries, we need another methodology that considers the updated information and keeps the accuracy of the result. In this paper we propose a novel approach based upon the duality and the marginal distribution to construct a histogram with very little time since the spatio-temporal histogram requires the data distribution defined by query predicates. We use data synopsis method in the dual space to construct spatio-temporal histograms. Our method is robust to changing data distributions during a certain period of time while the objects keep the linear movements. An additional feature of our approach supports the dynamic update incrementally and maintains the accuracy of the estimated result.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.1
• /
• pp.333-342
• /
• 2012

This paper presents a vowel onset point (VOP) detection method based on the human auditory system. This method maps the "perceptual" frequency scale, i.e. Mel scale onto a linear acoustic frequency, and then establishes a series of Triangular Mel-weighted Filter Bank simulate the function of band pass filtering in human ear. This nonlinear critical-band filter bank helps greatly reduce the data dimensionality, and eliminate the effect of harmonic waves to make the formants more prominent in the nonlinear spaced Mel spectrum. The sum of mel spectrum peaks energy is extracted as feature for each frame, and the instinct at which the energy amplitude starts rising sharply is detected as VOP, by convolving with Gabor window. For the single-word database which contains 12 vowels articulated with different kinds of consonants, the experimental results showed a good average detection rate of 72.73%, higher than other vowel detection methods based on short-time energy and zero-crossing rate.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.5
• /
• pp.486-496
• /
• 2013

Time synchronization in underwater environment is challenging due to high propagation delay and mobility of sensor nodes. Previous researches do not consider practical issues affecting on the accuracy of time synchronization such as high-channel access delay and relative position between sensor nodes. Also, those protocols using bidirectional message exchange shorten the network lifetime and decrease the network throughput because numerous transmission, reception and unnecessary overhearing can be occurred. Therefore, in our research, we suggest enhanced time synchronization based on features of underwater environment. It controls the instant of transmission by exploiting the feature of an oceanic movement and node deployment. Moreover, the protocol uses more accurate time information by removing channel access delay from the timestamp. The proposed scheme is also practical on the underwater sensor network requiring low-power consumption because the scheme conducts time-synchronization with smaller transmission and reception compared with previous works. Finally, simulation results show that the proposed protocol deceases time error by 2.5ms and 0.56ms compared with TSHL and MU-Sync respectively, reducing energy consumption by 68.4%.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.12
• /
• pp.573-580
• /
• 2017

Deep Neural Network (DNN) is a large layered neural network which is consisted of a number of layers of non-linear units. Deep Learning which represented as DNN has been applied very successfully in various applications. However, many issues in DNN have been identified through past researches. Among these issues, generalization is the most well-known problem. A Recent study, Dropout, successfully addressed this problem. Also, Dropout plays a role as noise, and so it helps to learn robust feature during learning in DNN such as Denoising AutoEncoder. However, because of a large computations required in Dropout, training takes a lot of time. Since Dropout keeps changing an inter-layer representation during the training session, the learning rates should be small, which makes training time longer. In this paper, using mutation operation, we reduce computation and improve generalization performance compared with Dropout. Also, we experimented proposed method to compare with Dropout method and showed that our method is superior to the Dropout one.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.18 no.1
• /
• pp.1-9
• /
• 2000

In the recent day, the necessity and the effective usage are increased rapidly, and it is applied in many other fields as well as in the field of ortho-photo map. In this study, we extract each objects on the aerial image and automatically classify graphic information to produce digital map using only digital ortho-image without particular drawing devices for producing digital map. For this purpose, we have applied a lot of the image processing techniques and fuzzy theory, classified outline and lane of road and building, and had each layer according to each feature. Especially, in the case of the building, the outer vector lines extracted by pixel unit at the building were very complex, but we have developed the program to be expressed by I-dimensional linear type between building corners. In the result of this study, we could not extract and recognize all of the object on the image all together, but we have got the error within 50cm using semi-automatic technique. Therefore, this method will be used effectively in producing 1/5,000 digital map.

• Journal of the Institute of Convergence Signal Processing
• /
• v.1 no.1
• /
• pp.15-22
• /
• 2000

The analysis of power spectrum based on linear AR model is applied widely to quantize the response of autonomic nerve noninvasively, In this paper, we estimate the power spectrum density for heartrate variability of the electrocadiogram and pulse wave for short term data(less than two minute), The time series of heart rate variability is obtained from the time interval(RRI, PPI) between the feature point of the electrocadiogram and pulse wave for normal person, The generated time series reconstructed into new time series through polynomial interpolation to apply to the AR mode. The power spectrum density for AR model is calculated by Burg algorithm, After applying AR model, the power spectrum density for heart rate variability of the electrocadiogram and the pulse wave is shown smooth spectrum power at the region of low frequence and high frequence, and that the power spectrum density of electrocadiogram and pulse wave has similar form for same subject.

• Journal of Energy Engineering
• /
• v.10 no.3
• /
• pp.272-277
• /
• 2001

An experimental study of jet impinging the non-isothermal heating surface with linear temperature gradient is conducted with the presentation of the turbulent flow characteristics and the heat transfer rate, represented by the Nusselt number. The jet Reynolds number ranges from 15,000 to 30,000, the temperature gradient of the plate is 2~4.2$^{\circ}C$/cm and the dimensionless nozzle to plate distance (H/D) is from 2 to 10. The results show that the peak of heat transfer rate occurs at the stagnation point, and the heat transfer rate decreases as the radial distance from the stagnation point increases. A remarkable feature of the heat transfer rate is the existence of the second peak. This is due to the turbulent development of the wall jet. Maximum heat transfer rate occurs when the axial distance from the nozzle to nozzle diameter (H/D) is 6 or 8. The heat transfer rate can be correlated as a power function of Prandtl number, Reynolds number, the dimensionless nozzle to plate distance (H/D) and temperature gradient (dT/dr). It has been found that the heat transfer rate increases with increasing turbulent intensity. The wall jet is influenced by temperature gradient and the effect becomes more important at higher radii.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.5
• /
• pp.2484-2498
• /
• 2017

Various types of social media make the people share their personal experience in different ways. In some social networking sites. Some users post their reviews, some users can support these reviews with comments, and some users just rate the reviews as kind of support or not. Unfortunately, there is rare explicit negative comments towards other reviews. This means if there is a link between two users, it must be positive link. Apparently, the negative link is invisible in these social network. Or in other word, the negative links are redundant to positive links. In this work, we first discuss the feature extraction from social media data and propose new method to compute the distance between each pair of comments or reviews on social media. Then we investigate whether we can predict negative links via regression analysis when only positive links are manifested from social media data. In particular, we provide a principled way to mathematically incorporate multi-facet data in a novel framework, Constructing Negative Links, CsNL to predict negative links for discovering the hidden information. Additionally, we investigate the ways of solution to general negative link predication problems with CsNL and its extension. Experiments are performed on real-world data and results show that negative links is predictable with multi-facet of social media data by the proposed framework CsNL. Essentially, high prediction accuracy suggests that negative links are redundant to positive links. Further experiments are performed to evaluate coefficients on different kernels. The results show that user generated content dominates the prediction performance of CsNL.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.4
• /
• pp.16-21
• /
• 2001

In this paper a virtual form of 3 dimensional face is synthesized from the two sheets of 2 dimensional photographs In this case two sheets of 2D face photographs, the front and the side photographs are used First of all a standard model for a general face is created and from this model the feature points which represents a construction of face are densely defined on part of ears. eyes, a nose and a lip but the other parts. for example, forehead, chin and hair are roughly determined because of flat region or the less individual points. Thereafter the side photograph is connected symmetrically on the left and right sides of the front image and it is gradually synthesized by use of affine transformation method. In order to remove the difference of color and brightness from the junction part, a linear interpolation method is used. As a result it is confirmed that the proposed model which general model of a face can be obtain the 3D virtual image of the individual face.