• Journal of the Korean Data and Information Science Society
• /
• v.26 no.2
• /
• pp.495-504
• /
• 2015

The Gaussian geostatistical model has been widely used for modeling spatial data. However, this model suffers from a severe difficulty in computation because inference requires to invert a large covariance matrix in evaluating log-likelihood. In addressing this computational challenge, three strategies have been employed: likelihood approximation, lower dimensional space approximation, and Markov random field approximation. In this paper, we reviewed statistical approaches attacking the computational challenge. As an illustration, we also applied integrated nested Laplace approximation (INLA) technology, one of Markov approximation approach, to real data to provide an example of its use in practice dealing with large spatial data.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.2 no.2
• /
• pp.131-139
• /
• 1999

An easy and efficient method is proposed for a computation of reliability of preprocessing filters in the fire control system when the sensor data are frequently unreliable depending on the operation environment. It computes state transition probability matrix after modeling filter states as a Markov process, and computing false alarm and detection probability of each filter state under the given sensor failure probability. It shows that two important indices such as distributed state probability and error variance can be derived easily for a reliability assessment of the given sensor fusion system.

• Korean Journal of Remote Sensing
• /
• v.17 no.4
• /
• pp.335-344
• /
• 2001

Texture is extensively studied in a variety of image processing applications such as image segmentation and classification because it is an important property to perceive regions and surfaces. This paper focused on the analysis and synthesis of textured single and multiband images using Markov Random Field model considering the existent spatial correlation. Especially, for multiband images, the cross-channel correlation existing between bands as well as the spatial correlation within band should be considered in the model. Although a local interaction is assumed between the specified neighboring pixels in MRF models, during the maximization process, short-term correlations among neighboring pixels develop into long-term correlations. This result in exhibiting phase transition. In this research, the role of temperature to obtain the most probable state during the sampling procedure in discrete Markov Random Fields and the stopping rule were also studied.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.488-488
• /
• 2015

Intermittent streamflow is common phenomenon in arid and semi-arid regions. To manage water resources of intermittent streamflows, stochactic simulation data is essential; however the seasonally stochastic modeling for intermittent streamflow is a difficult task. In this study, using the periodic Markov chain model, we simulate intermittent monthly streamflow for occurrence and the periodic gamma autoregressive and copula models for amount. The copula models were tested in a previous study for the simulation of yearly streamflow, resulting in successful replication of the key and operational statistics of historical data; however, the copula models have never been tested on a monthly time scale. The intermittent models were applied to the Colorado River system in the present study. A few drawbacks of the PGAR model were identified, such as significant underestimation of minimum values on an aggregated yearly time scale and restrictions of the parameter boundaries. Conversely, the copula models do not present such drawbacks but show feasible reproduction of key and operational statistics. We concluded that the periodic Markov chain based the copula models is a practicable method to simulate intermittent monthly streamflow time series.

• Journal of the Society of Naval Architects of Korea
• /
• v.40 no.5
• /
• pp.43-52
• /
• 2003

Two methods which estimate manoeuvring derivatives in the model of hydrodynamic force and moment acting on a manoeuvring ship using sea trial data were compared. One is the widely used parameter estimation method by using the Extended Kalman Filter (EKF), which estimates state variables of linearized state space model at every instant after dealing with the coefficients as the augmented state variables. The other one is the Estimation-Before-Modeling (EBM) technique, so called the two-step method. In the first step, hydrodynamic force of which dynamic model is assumed the third-order Gauss-Markov process is estimated along with motion variables by the EKF and the modified Bryson-Frazier smoother. Then, in the next step, manoeuvring derivatives are identified through the regression analysis. If the exact structure of hydrodynamic force could be known, which was an ideal case, the EKF method would be regarded as being more superior compared to the EBM technique. However the EBM technique was more robust than the EKF method from a realistic point of view where the assumed model structure was slightly different from the real one.

• Journal of KIISE:Information Networking
• /
• v.29 no.6
• /
• pp.674-684
• /
• 2002

Anomaly detection techniques have teen devised to address the limitations of misuse detection approach for intrusion detection. An HMM is a useful tool to model sequence information whose generation mechanism is not observable and is an optimal modeling technique to minimize false-positive error and to maximize detection rate, However, HMM has the short-coming of login training time. This paper proposes an effective HMM-based IDS that improves the modeling time and performance by only considering the events of privilege flows based on the domain knowledge of attacks. Experimental results show that training with the proposed method is significantly faster than the conventional method trained with all data, as well as no loss of recognition performance.

• Nuclear Engineering and Technology
• /
• v.51 no.2
• /
• pp.444-452
• /
• 2019

To assess the availability of a nuclear power plant's dynamic systems, it is necessary to consider the impact of dynamic interactions, such as components, software, and operating processes. However, there is currently no simple, easy-to-use tool for assessing the availability of these dynamic systems. The existing method, such as Markov chains, derives an accurate solution but has difficulty in modeling the system. When using conventional fault trees, the reliability of a system with dynamic characteristics cannot be evaluated accurately because the fault trees consider reliability of a specific operating configuration of the system. The dynamic reliability graph with general gates (DRGGG) allows an intuitive modeling similar to the actual system configuration, which can reduce the human errors that can occur during modeling of the target system. However, because the current DRGGG is able to evaluate the dynamic system in terms of only reliability without repair, a new evaluation method that can calculate the availability of the dynamic system with repair is proposed through this study. The proposed method extends the DRGGG by adding the repair condition to the dynamic gates. As a result of comparing the proposed method with Markov chains regarding a simple verification model, it is confirmed that the quantified value converges to the solution.

• International Journal of Highway Engineering
• /
• v.4 no.4 s.14
• /
• pp.1-12
• /
• 2002

Pavement Management System has a special purpose that the rehabilitation strategy applied on pavement should be executable in view of technical and economical point after new pavement open to the traffic. To achieve that purpose, a reliable pavement performance prediction model should be embeded in the system. The object of this study is to develop a probabilistic pavement performance prediction model for evaluating asphalt pavements based on the Markov chain concept. In this paper, methodology of the Markov chain modeling principle is explained, and the application of this model to asphalt pavement is described. As the results, transition matrics for predicting asphalt pavement performance are obtained, and also performance life is estimated quantitatively by this system.

• International Journal of Contents
• /
• v.5 no.4
• /
• pp.88-93
• /
• 2009

It is important for communication networks to possess the capability to overcome failures and provide survivable services. We address modeling and analysis of performability affected by both performance and availability of system components for a token ring network under failure and repair conditions. Stochastic reward nets (SRN) is an extension of stochastic Petri nets and provides compact modeling facilities for system analysis. In this paper, hierarchical SRN modeling techniques are used to overcome state largeness problem. The upper level model is used to compute availability and the lower level model captures the performance. And Normalized Throughput Loss (NTL) is obtained for the composite ring network for each node failures occurrence as a performability measure. One of the key contributions of this paper constitutes the Petri nets modeling techniques instead of complicate numerical analysis of Markov chains and easy way of performability analysis for a token ring network under SRN reward concepts.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.3B
• /
• pp.277-284
• /
• 2011

The limitations of existing Markov chain model for reproducing extreme rainfalls are a known problem, and the problems have increased the uncertainties in establishing water resources plans. Especially, it is very difficult to secure reliability of water resources structures because the design rainfall through the existing Markov chain model are significantly underestimated. In this regard, aims of this study were to develop a new daily rainfall simulation model which is able to reproduce both mean and high order moments such as variance and skewness using a piecewise Kernel-Pareto distribution. The proposed methods were applied to summer and fall season rainfall at three stations in Han river watershed in Korea. The proposed Kernel-Pareto distribution based Markov chain model has been shown to perform well at reproducing most of statistics such as mean, standard deviation and skewness while the existing Gamma distribution based Markov chain model generally fails to reproduce high order moments. It was also confirmed that the proposed model can more effectively reproduce low order moments such as mean and median as well as underlying distribution of daily rainfall series by modeling extreme rainfall separately.