• The Journal of the Acoustical Society of Korea
• /
• v.23 no.8
• /
• pp.590-596
• /
• 2004

This paper Presents a new self-adaptation algorithm based on maximum a posteriori (MAP) eigenvoice for Korean connected digit recognition. The proposed MAP eigenvoice is developed by introducing a probability density model for the eigenvoice coefficients. The Proposed approach provides a unified framework that incorporates the Prior model into the conventional eigenvoice estimation. In self-adaptation system we use only one adaptation utterance that will be recognized, we use MAP eigenvoice that is most robust adaptation. In series of self-adaptation experiments on the Korean connected digit recognition task. we demonstrate that the performance of the proposed approach is better than that of the conventional eigenvoice algorithm for a small amount of adaptation data.

• Journal of the Korea Computer Graphics Society
• /
• v.3 no.1
• /
• pp.37-47
• /
• 1997

This paper presents the algorithms for generating ambient traffic in driving simulation. Each ambient car is modeled as an autonomous agent that obeys the traffic rules by sensing the traffic lights, road signs, lanes, and other cars around. The algorithm is localized to the area where the car driven by the participant is currently located. Therefore the complexity of the algorithm does not depend on the size of the road network, allowing a huge environment to be simulated with no extra overhead. To avoid monotony, we produce artificial fluctuations in the behavior by employing various forms of probability distribution functions. The resulting behavior of the ambient cars is quite realistic. Experiments indicate that it is hard to tell whether an ambient car is computer-controlled or human-controlled.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.33-33
• /
• 2011

현재 우리나라 기상청에서는 단기, 중기 및 장기 예보자료를 생산하고 있으나, 이들 자료는 단순히 일기 예보에 치중되어 생산되고 있어 강우-유출해석에 직접 적용하기에는 시 공간 해상도가 크고 정량적 강수예측의 정확도가 미흡하다. 이에 기상 및 수자원분야에서는 정확도 개선을 위해서 관측강우와 예측강우의 비교 분석을 통해 편차를 산정하여 예측강수를 보정하는 기법을 적용하고 있다. 다만, 기존의 편차보정방법은 보정인자로 강수량만을 고려하기 때문에 정확도 개선에는 한계가 존재한다. 따라서 본 연구에서는 수자원분야의 수치예보자료의 정확도를 향상시키기 위해 규모, 발생영역에 대한 강수의 특성을 고려한 강수예측자료의 편차보정 방법을 제안하고 이를 강우-유출모델에 적용하여 개선정도를 평가하고자 한다. 이에 적용유역을 춘천댐상류유역으로 선정하고 국내 기상청의 RDAPS(Regional Data Assimilation and Prediction System)수치예보자료, 지점강우자료, radar자료의 수문기상자료와 지형자료를 수집하였다. 화천, 평화의 댐 일부 미계측유역의 관측자료로 radar자료를 이용하였다. 이상의 자료를 토대로 강우강도 및 규모, 영향범위를 고려한 예측강우의 편차를 산정하여 RDAPS 수치예보자료의 정확도를 개선하고 평가하였다. 이는 해당 유역뿐만 아니라 주변 유역의 정보를 이용하여 예측강우의 발생위치에 대한 오차를 고려한 방법으로, 각 영역별로 예측강우의 편차보정계수를 산정하여 적용하였다. 또한, 이전시간대의 강우 편차에 대한 오차를 줄이기 위해 정규분포방법을 이용한 Ensemble 편차보정계수를 산정하고 최근 생산된 수치예보자료에 적용하여 확률예측강우를 산정하였다.

• The KIPS Transactions:PartC
• /
• v.12C no.6 s.102
• /
• pp.793-798
• /
• 2005

We focus on the RP (Rendezvous Point) system model in the multicast network based on BcN (Broadband Convergence Network) integrating broadcasting, telecommunication and Internet with one. Based on the condition multiple queues with different service and single server, when the arrivals tome in group with the site of the group geometrically distributed, we define the relationship between incoming arrival rate and corresponding buffer size. We also investigate the Profit according to both Service Provider and Network Operator Then we make a decision whether a new service request is accepted or not based on given interning rate range.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.5
• /
• pp.1119-1129
• /
• 1998

This paper proposes a method to estimate the time to execute a hypermedia scenario. The method builds a PNDT(Petri Net with probability Distribution Table) model for the flow of the scenario, then performs Monte Carlo Simulation on the PNDT model to estimate the duration time from the start to the end of the scenario. The estimated times to execute scenarios are useful to both information providers and users. A user can make a better time plan referring to them, and an information provider can make the most efficient scenario-presentation schedule for users referring to them. However, no research results regarding this subject have been reported yet.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.5 s.72
• /
• pp.705-712
• /
• 2004

To investigate the mechanical behavior of pin cups created from steel pipes embedded in concrete, test blocks were made and load tests were conducted. The pin was inserted in the pin cup at various depths and was loaded in a lateral direction at a constant displacement speed. A formula, which took into consideration the deformation and failure shapes of the pin cups, observed during and after the tests, was used to estimate its mechanical behavior. The test values were determined to be relevant to the formula and were distributed within a reasonably narrow range. The mean and the 95% survival probability value of maximum resistance were determined by factoring the formula at 1.01 and 0.92, respectively.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.412-417
• /
• 2005

The Internet has been widely applied to the remote control system. The network-based control system, however, has a random time delay and an inherent weak point of the network, when the data ate transmitted. The network delay may result in performance degradation or even system instability in teleoperation. In this paper a prediction model of network delay using TSK (Takagi-Sugeno-Kang) fuzzy model is presented. An adaptive scheme is developed to update the prediction model according to the current network status. The prediction model is applied to the control of an Internet-based mobile robot to show its usefulness. In the computer simulation the TSK Prediction model of network delay is proven superior to the conventional algorithms.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.4
• /
• pp.18-23
• /
• 1996

Fluctuation statistics of scattering strength are not only important because they impact the performance of active sonar systems, but also because they may provide insight into the major scattering process. In this article, analysis of the statistical characteristics of bottom backscattering, measured in shallow water, are presented. The slowly moving experimental sonar was operated at 30kHz to gather data over the bottom. Spatial and temporal correlation functions of the signal amplitudes were measured. The distribution function and probability of false alarm function of the detected envelope of widebeam and narrowbeam signals were measured. An attempt was made to compare the results with existing theoretical models. The result suggests that the statistical characteristics of bottom backscattering fluctuation of moving source is differ from that of fixed source.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.280-280
• /
• 2016

특정 자료의 시간의 흐름에 따른 예측치를 추정하는 방법으로 AR Model 즉, 자기회귀모형이 많이 사용되고 있다. AR Model은 변수의 현재 값을 과거 값의 함수로 나타내게 되는데, 이런 시계열 분석 모델을 사용할 때 매개변수의 추정 과정이 필수적으로 요구된다. 일반적으로 매개변수를 추정하는 방법에는 확률적근사법(stochastic approximation), 최소제곱법(method of least square), 자기상관법(method of autocorrelation method), 최우도법(method of maximum likelihood) 등이 있다. AR Model에서 가장 많이 사용되는 최우도법은 표본크기가 충분히 클 때 가장 효율적인 방법으로 평가되지만 수치적으로 해를 구하는 과정이 복잡한 경우가 많으며, 해를 구하지 못하는 어려움이 따르기도 한다. 또한 표본 크기가 작을 때 일반적으로 잘 일치하지 않은 결과를 얻게 된다. 우리나라의 강우, 유량 등의 자료는 자료의 수가 적은 경우가 많기 때문에 최우도법을 통한 매개변수 추정 시 불확실성이 내재되어있지만 그것을 정량적으로 제시하는데 한계가 있다. 본 연구에서는 AR Model의 매개변수 추정 시 Bayesian 기법으로 매개변수의 사후분포(posterior distribution)를 제공하여 매개변수의 불확실성 구간을 정량적으로 표현하게 됨으로써, 시계열 분석을 통해 보다 신뢰성 있는 예측치를 얻을 수 있으리라 판단된다.

• Journal of the Korean Geosynthetics Society
• /
• v.17 no.4
• /
• pp.103-108
• /
• 2018

The fragility curves for CFRD dams are derived in this study for probabilistic damage estimation as a function of a ground motion intensity. The dam crest settlement, which is a widely used damage index, is used for minor, moderate, and extensive damage states. The settlement is calculated from nonlinear dynamic numerical simulations. The accuracy of the numerical model is validated through comparison with a centrifuge test. The fragility curve is represented as a log normal distribution function and presented as a function of the peak ground acceleration. The fragility curves developed in this study can be utilized for real time assessment of the damage of dams.