• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.65-67
• /
• 2018

We introduce a system that detects smoking in a specific area. The proposed system is implemented on a wireless ad hoc network consisting of Raspberry Pi boards. It is more economical owing to low-cost device than commercial smoking monitoring system and is scalable than the existing system with single Raspberry Pi. In this paper, the probability density function of carbon monoxide concentration during smoking and non-smoking is approximated as Gauusian distribution, respectively, using data measured from sensors for a long time. Based on this, a maximum likelihood detection technique is adopted to estimate the smoking status by observing the concentration of carbon monoxide. We aim at improving the reliability by estimating the smoking status using the collected values from multiple sensors connected to the ad hoc network.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.4
• /
• pp.383-389
• /
• 2011

In this paper, a statistical calibration method is proposed in order to identify the variability of complex modulus for a rubber material due to operational temperature and experimental/model errors. To describe temperature- and frequency-dependent material properties, a fractional derivative model and a shift factor relationship are used. A likelihood function is defined as a product of the probability density functions where experimental values lie on the model. The variation of the fractional derivative model parameters is obtained by maximizing the likelihood function. Using the proposed method, the variability of a synthetic rubber material is estimated and applied to a rubber mount problem. The dynamic characteristics of the rubber mount are calculated using a finite element model of which material properties are sampled from Monte Carlo simulation. The calculated dynamic stiffnesses show very large variation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.3
• /
• pp.149-154
• /
• 2014

The cross-correntropy concept can be expressed with inner products of two different probability density functions constructed by Gaussian-kernel density estimation methods. Blind algorithms based on the maximization of the cross-correntropy (MCC) and a symbol set of randomly generated N samples yield superior learning performance, but have a huge computational complexity in the update process at the aim of weight adjustment based on the MCC. In this paper, a method of reducing the computational complexity of the MCC algorithm that calculates recursively the gradient of the cross-correntropy is proposed. The proposed method has only O(N) operations per iteration while the conventional MCC algorithms that calculate its gradients by a block processing method has $O(N^2)$. In the simulation results, the proposed method shows the same learning performance while reducing its heavy calculation burden significantly.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.5
• /
• pp.525-530
• /
• 2006

This paper presents an efficient input variable selection method using both fixed-point independent component analysis(FP-ICA) and adaptive partition mutual information(AP-MI) estimation. FP-ICA which is based on secant method, is applied to quickly find the independence between input variables. AP-MI estimation is also applied to estimate an accurate dependence information by equally partitioning the samples of input variable for calculating the probability density function(PDF). The proposed method has been applied to 2 problems for selecting the input variables, which are the 7 artificial signals of 500 samples and the 24 environmental pollution signals of 55 samples, respectively The experimental results show that the proposed methods has a fast and accurate selection performance. The proposed method has also respectively better performance than AP-MI estimation without the FP-ICA and regular partition MI estimation.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.2
• /
• pp.97-102
• /
• 2005

Acoustic experiments were performed with Zostera marina to study the characteristics of backscattering of seagrass living in the bottom interface. Field experiments were conducted in the Dongdae man, Namhae for day and night to consider the effects of air-bubble from photosynthesis of seagrass. The multi-frequency (30$\~$120 kHz) responses were measured and the distributions of back scattering strength due to the movement of seagrass were Presented by PDF (probability density function) at 120 120 kHz. The results were shown both the frequency dependence and diurnal variation of the backscattering strength between day and night. This diurnal variation may be caused by the amount of oxygen in dissolved bubbles formed by Photosynthesis of seagrass.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.11 no.1
• /
• pp.141-152
• /
• 2007

Considering uncertainties of random input data, it is more reasonable to use probabilistic method than the conventional deterministic method for the design of structures or for the assessment of the responses of structures, which are designed as safe even under extreme loads. Therefore, to assess the quantitative effects of the constructed cable stayed bridge by the input random variables, a sensitivity analysis is studied. Using perturbation method, an analysis program is developed for the iterative probabilistic finite element analyses and sensitivity analyses of the cable stayed bridge, except the initial shape analysis. Monte-Carlo Simulations were used for the verification of the developed program. The results of sensitivity analysis shows the governing effects of external loads. Because the results also provide the sensitive effects of the stiffness of members and the magnitudes of prestressing force of cables, the developed

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.423-426
• /
• 2007

클러스터링은 주어진 데이터 포인트들을 주어진 개수의 그룹으로 나누는 비지도 학습의 한 방법이다. 클러스터링의 방법 중 하나로 널리 알려진 퍼지 클러스터링은 하나의 포인트가 모든 클러스터에 서로 다른 정도로 소속될 수 있도록 함으로써 각 포인트가 하나의 클러스터에만 속할 수 있도록 하는 K-means와 같은 방법에 비해 자연스러운 클러스터 형태의 유추가 가능하고, 잡음에 강한 장점이 있다. 이 논문에서는 기존의 퍼지 클러스터링 방법 중 소속도(membership)와 전형성(typicality)을 동시에 계산해 낼 수 있는 Possibilistic Fuzzy C-Means (PFCM) 방법에 Gath-Geva (GG)의 방법 을 적용하여 PFCM을 확장한다. 제안한 방법은 PFCM의 장점을 그대로 가지면서도, GG의 거리 척도에 의해 클러스터들 사이의 경계를 강조함으로써 분류 목적에 적합한 소속도를 계산할 수 있으며, 전형성은 가우스 형태의 분포에서 생성된 포인트들의 분포 함수를 정확하게 모사함으로써 확률 밀도 추정의 방법으로도 사용될 수 있다. 또한 GG 방법은 Gustafson-Kessel 방법과 달리 클러스터에 포함된 포인트의 개수가 확연히 차이 나는 경우에도 정확한 결과를 얻을 수 있다는 사실을 실험 결과를 통해 확인할 수 있었다.

• Journal of the Korean Geographical Society
• /
• v.34 no.5
• /
• pp.473-491
• /
• 1999

세균성 이질은 국내 제1종 법정 전염병으로 분류되어 관리되고 있는 질환으로서 1998년 이후 그 발병 사례가 급속히 증가하고 있다. 본 연구는 1999년 3월 부산시 사상구에서 집단 발병한 세균성 이질을 대상으로 하여, 각 환자들의 발병 시점과 장소의 분포패턴에 대한 지리학적 고찰을 목적으로 한다. 환자분포의 특징적 공간패턴과 그들의 시계열적 확산 양상 등을 탐색하기 위한 방법론은 보건지리학과 지도학 및 공간통계학에 기반을 둔 공간분석기법을 중심으로 설정하였다. 분석자료는 해당 지역의 수치지형도, 지적도, 인구 센서스 자료를 포함한 GIS 데이터베이스로 구축되었다. 인구분포를 감안한 밀도구분도를 바탕으로 개별환자의 위치자료와 동 단위로 집계된 자료를 자료의 형태에 따라 분석기법을 달리하였으며, 환자 발생 밀도, 상대적 위험지수 등을 지도화하여 역학자료의 시각적 통계적 분석을 수행하였다. 환자분포의 공간적 중심위치와 분산의 변화 등 기술적 통계분석과 함께 제1차 공간속성을 커널추정법으로 찾아보았다. 이와 더불어 ‘공간적 의존성’과 관련된 제2차 공간속성은 K-함수와 시뮬레이션을 통해 분석하여 군집성 등이 통계적으로 확인되었다. 본 연구를 통해 역학조사시 GIS의 활용사례가 제시되었으며, 모집단 인구를 고려한 확률지도 작성 기법과 다양한 데이터 가시화 방법, 그리고 시계열별 발생 환자들의 지리적 변이를 분석 하는데 따르는 문제들이 논의되었다.

• Journal of Korea Water Resources Association
• /
• v.40 no.2 s.175
• /
• pp.159-170
• /
• 2007

This study evaluated the parameters of Clark unit hydrograph (UH) estimated using the rainfall-runoff measurements and evaluated their variability. This also includes the quantification of basin and meteorological factors using probability density functions, selection of storm events with mean affecting factors, and derivation of average parameters of the Clark UH from storm events selected. Summarizing the results from this procedure are as follows. (1) It is not easy to avoid much uncertainty on the decision of runoff characteristics (that is, the concentration time and storage coefficient) even with some rainfall-runoff events are available. (2) As the distribution function of concentration time is very skewed, a simple arithmetic mean may lead a biased estimate. That is, the arithmetic mean based on the normal distribution can not be representative anymore. The mode may well be the representative in this case. On the other hand, the storage coefficient shows a symmetric distribution function, so the arithmetic mean may be used use for its representative. For the basin in this study, the concentration time in this study is estimated to be about 7 hours, and the storage coefficient about 22 hours.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.6B
• /
• pp.599-606
• /
• 2010

Due to the short period of precipitation data in Korea, the uncertainty of drought analysis is inevitable from a point frequency analysis. So it is desired to introduce a regional drought frequency analysis. This study first extracted drought characteristics from 3-month and 12-month moving average rainfalls which represent short and long-term droughts, respectively. Then, the homogeneous regions were distinguished by performing a principal component analysis and cluster analysis. The Korean peninsula was classified into five regions based on drought characteristics. Finally, this study applied the bivariate frequency analysis using a kernel density function to quantify the regionalized drought characteristics. Based on the bivariate drought frequency curves, the drought severities of five regions were evaluated for durations of 2, 5, 10, and 20 months, and return periods of 5, 10, 20, 50, and 100 years. As a result, the largest severity of drought was occurred in the Lower Geum River basin, in the Youngsan River basin, and over in the southern coast of Korea.