• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04a
• /
• pp.841-843
• /
• 2004

비디오 스트링의 표준이라 할 수 있는 MPEG은 데이터 발생량의 변화가 심한 군집성(bursty) 트래픽으로 망의 대역폭을 효율적으로 사용하는 전송방식을 구현하기가 매우 어렵다. 본 연구에서는 최소한의 품질을 보장하면서도 망 자원의 효율적 이용을 위하여 2계층 구조의 새로운 대역폭 할당 기법을 제안하였다. 사용자에게 최소한의 품질을 보장하면서 망에 대역폭의 여유가 있는 경우 보다 고품질의 서비스가 가능토록 하는 방안으로 ATM망의 CBR 클래스와 VBR 클래스를 복합적으로 사용하는 방법을 제안하였다. 이의 구현을 위하여 2계층 구조의 MPEG 부호화기를 설계하였고 모의실험으로 기존의 단일 계층 CBR 클래스와 비교 평가하였다.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.31-39
• /
• 2006

In many cases, we frequently get a desired information based on the appropriate statistical analysis of collected data sets. Lots of statistical theory rely on the assumption of the normality of the data. In this paper, we compare the empirical power of some normality tests including sample entropy quantity. Monte carlo simulation is conducted for the calculation of empirical power of considered normality tests by varying sample sizes for various distributions.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1999.10a
• /
• pp.47-51
• /
• 1999

가스터빈 발전소의 제어시스템을 개조할 때, 설계자가 구성한 디지털 제어시스템의 동작특성을 검증하기 위해서는 시뮬레이터를 구축하여 모의실험을 할 필요가 있다. 본 논문에서는 시뮬레이터를 구성하는데 필요한 구조와 실제 시뮬레이션 한 방법 및 결과에 대해서 언급하고자 한다. 이를 위해 실제 발전소에서 운전데이터를 취득하였고, 시뮬레이션 결과 현장과 유사한 결과를 보여줌으로써 시뮬레이터 구성의 타당성 및 필요성을 확인시켜 주었다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.1
• /
• pp.1-8
• /
• 2001

본 논문에서는 정규화된 데이터 공간과 가우스함수에 의한 산 함수 형성 그리고 형성된 산의 기울기를 이용한 산봉우리 붕괴를 특징으로 하는 개선된 산 클러스터링 방법을 제안한다. 이 개선된 방법은 기존의 Yager 등에 의하여 제안된 방법이 조정해야 하는 매개변수가 3개이고 발견된 클러스터 중심 주위에 원치 않는 다른 중심이 발생할 수 있는데 반하여 단지 하나의 매개변수 $\omega$의 조정으로 더욱 타당한 중심을 찾아내는 점에서 유용하다 할 수 있다. 또한 매개변수 $\omega$에 대한 적절한 선정 방법을 제시하고, 수치 자료에 대한 컴퓨터 모의실험을 통하여 개선된 산 클러스터링 방법의 유용성을 입증한다.

• Journal of the Korean Data and Information Science Society
• /
• v.1
• /
• pp.47-57
• /
• 1990

A method which determines the number of replications in the simulation is proposed, particularly for small-sample comparison of estimators. This method takes the smallest number of replications that makes the difference of mean square errors be statistically significant and provides an efficient algorithm for calculating the standard error of the mean square error. Two examples are illustrated, the first one is on comparison of mean and median ; the second, the Kaplan-Meier type and Buckley-James type estimators of a quantile function with censored data.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2015.11a
• /
• pp.246-249
• /
• 2015

본 연구에서는 해운대 해수욕장 연안정비사업 전후의 지형변화에 따른 이안류 발생 변화에 대해 분석하였다. 본 연구에서는 Matlab GUI (Graphical User Interface)로 개발한 HAECUM (HAE(海) Current Model)을 통하여 해운대 해수욕장의 해빈류를 모의 하였다. 분석에 사용된 데이터는 이상적인 경우와 실제 이안류가 발생한 상황에 대해 각각 적용하였으며, 상호비교의 효율성을 위하여 해빈류 장미도를 통해 비교 분석 하였다. 본 연구를 통해 해운대 해수욕장의 연안정비사업에 따른 해빈류 및 이안류에 대한 상관성 도출하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.1069-1070
• /
• 2009

본 논문에서는 실시간 학습기능을 갖는 다층 신경회로망과 주어지는 인터넷 트래픽 데이터에서의 IDC(Index dispersion for count) 정보를 이용하여 인터넷에서의 공격(침입)을 탐지 예측 할 수 있는 모델을 제안하고 컴퓨터 모의실험을 통한 결과를 보인다.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.10 no.6
• /
• pp.155-163
• /
• 2010

The past researches on flood inundation simulation mainly focused on development of numerical models based on unstructured mesh networks to improve model performances. However, despite the accurate simulation results, such models are not suitable for real-time flood inundation forecasting due to a huge computational burden in terms of geographic data processing. In addition, even though various types of vector and raster data are available to be compatible with flood inundation models for post-processes such as flood hazard mapping and flood inundation risk analysis, the unstructured mesh-based models are not effective to fully use such information due to data incommensurability. Therefore, this study aims to develop a raster-based two-dimensional inundation model; it guarantees computational efficiency because of direct application of DEM for flood inundation modeling and also has a good compatibility with various types of raster data, compared to a commercial model such as FLUMEN. We applied the model to simulate the BaekSan levee break in the Nam river during a flood period from August 10 to 13, 2002. The simulation results showed a good agreement with the field-surveyed inundation area and were also very similar with results from the FLUMEN. Moreover, the model provided physically-acceptable velocity vectors with respect to inundating and returning flows due to the difference of water level between channel and lowland.

• Journal of Korea Water Resources Association
• /
• v.56 no.9
• /
• pp.563-574
• /
• 2023

Recently, climate change has affected functional responses of river basins to meteorological variables, emphasizing the importance of rainfall-runoff simulation research. Simultaneously, the growing interest in machine learning has led to its increased application in hydrological studies. However, it is not yet clear whether machine learning models are more advantageous than the conventional conceptual models. In this study, we compared the performance of the conventional GR6J model with the machine learning-based Random Forest model across 38 basins in Korea using both gauged and ungauged basin prediction methods. For gauged basin predictions, each model was calibrated or trained using observed daily runoff data, and their performance was evaluted over a separate validation period. Subsequently, ungauged basin simulations were evaluated using proximity-based parameter regionalization with Leave-One-Out Cross-Validation (LOOCV). In gauged basins, the Random Forest consistently outperformed the GR6J, exhibiting superiority across basins regardless of whether they had strong or weak rainfall-runoff correlations. This suggest that the inherent data-driven training structures of machine learning models, in contrast to the conceptual models, offer distinct advantages in data-rich scenarios. However, the advantages of the machine-learning algorithm were not replicated in ungauged basin predictions, resulting in a lower performance than that of the GR6J. In conclusion, this study suggests that while the Random Forest model showed enhanced performance in trained locations, the existing GR6J model may be a better choice for prediction in ungagued basins.

• The Korean Journal of Applied Statistics
• /
• v.27 no.7
• /
• pp.1171-1185
• /
• 2014

Rhipe that integrates R and Hadoop environment, made it possible to process and analyze massive amounts of data using a distributed processing environment. In this paper, we implemented multiple regression analysis using Rhipe with various data sizes of actual data and simulated data. Experimental results for comparing the computing speeds of pseudo-distributed and fully-distributed modes for configuring Hadoop cluster, showed fully-distributed mode was more fast than pseudo-distributed mode and computing speeds of fully-distributed mode were faster as the number of data nodes increases. We also compared the performance of our Rhipe with stats and biglm packages available on bigmemory. The results showed that our Rhipe was more fast than other packages owing to paralleling processing with increasing the number of map tasks as the size of data increases.