• Proceedings of the KSME Conference
• /
• 2003.11a
• /
• pp.1846-1851
• /
• 2003

Approximate optimization has become popular in engineering field such as MDO and Crash analysis which is time consuming. To accomplish efficient approximate optimization, accuracy of approximate model is very important. As surrogate model, Kriging have been widely used approximating highly nonlinear system . Because Kriging employs interpolation method, it is adequate for deterministic computer simulation. Because there are no random errors and measurement errors in deterministic computer simulation, instead of classical DOE ,space filling experiment design which fills uniformly design space should be applied. In this work, various space filling designs such as maximin distance design, maximum entropy design are reviewed. And new design improving maximum entropy design is suggested and compared.

• Journal of Communications and Networks
• /
• v.10 no.1
• /
• pp.89-97
• /
• 2008

Anomaly detection systems playa significant role in protection mechanism against attacks launched on a network. The greatest challenge in designing systems detecting anomalous exploits is defining what to measure. Effective yet simple, Shannon entropy metrics have been successfully used to detect specific types of malicious traffic in a number of commercially available IDS's. We believe that Renyi entropy measures can also adequately describe the characteristics of a network as a whole as well as detect abnormal traces in the observed traffic. In addition, Renyi entropy metrics might boost sensitivity of the methods when disambiguating certain anomalous patterns. In this paper we describe our efforts to understand how Renyi mutual information can be applied to anomaly detection as an offline computation. An initial analysis has been performed to determine how well fast spreading worms (Slammer, Code Red, and Welchia) can be detected using our technique. We use both synthetic and real data audits to illustrate the potentials of our method and provide a tentative explanation of the results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.2
• /
• pp.200-205
• /
• 2008

For data analysis, fuzzy entropy is introduced as the measure of fuzziness, similarity measure is also constructed to represent similarity between data. Similarity measure between fuzzy membership functions is constructed through distance measure, and the proposed similarity measure are proved. Application of proposed similarity measure to the example of reliable data selection is also carried out. Application results are compared with the previous results that is obtained through fuzzy entropy and statistical knowledge.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.283-284
• /
• 2006

In this paper, we propose a novel segmentation and classification method using texture features for the document image. First, we extract the local entropy and then segment the document image to separate the background and the foreground using the Otsu's method. Finally, we classify the segmented regions into each component using PCA(principle component analysis) algorithm based on the texture features that are extracted from the co-occurrence matrix for the entropy image. The entropy-based segmentation is robust to not only noise and the change of light, but also skew and rotation. Texture features are not restricted from any form of the document image and have a superior discrimination for each component. In addition, PCA algorithm used for the classifier can classify the components more robustly than neural network.

• Structural Engineering and Mechanics
• /
• v.3 no.2
• /
• pp.135-144
• /
• 1995

Since structural systems may fail in any one of several failure modes, computation of system reliability is always difficult. A method using numerical quadrature for computing structural system reliability with either one or more than one failure mode is presented in this paper. Statistically correlated safety margin equations are transformed into a group of uncorrelated variables and the joint density function of these uncorrelated variables can be generated by using the Maximum Entropy Method. Structural system reliability is then obtained by integrating the joint density function with the transformed safety domain enclosed within a set of linear equations. The Gaussian numerical integration method is introduced in order to improve computational accuracy. This method can be used to evaluate structural system reliability for Gaussian or non-Gaussian variables with either linear or nonlinear safety boundaries. It is also valid for implicit safety margins such as computer programs. Both the theory and the examples show that this method is simple in concept and easy to implement.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.24 no.9
• /
• pp.671-678
• /
• 2012

The objective of this study is to analyze theoretically the performance of an open air-cycle refrigeration system in which environmental concerns increase. The pressure ratio of the external compressor and efficiencies of the components that compose of the system are selected as important parameters. As the pressure ratio of the external compressor increases, the pressure ratio of the ACM compressor is determined high, the refrigerating temperature and capacity increase, the COP decreases, and the total entropy production rate increases. The effect of heat exchanger effectiveness and turbine efficiency on the performance are greater than that of the ACM compressor efficiency. Also the performance of the air-cycle refrigeration system with two heat exchangers has been enhanced like high COP and low total entropy production rate, compared to the system with one heat exchanger.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2004.10a
• /
• pp.27-35
• /
• 2004

In this paper a method for finding solution of acoustic, vortex and entropy wave-equations in a cylindrical tube with variable section area was suggested under the consideration of that the high frequency instability in a rocket engine combustion chamber is an acoustic phenomena, which is coupled with combustion reaction, and that a combustion chamber and exhaust nozzle are usually shaped cylindrically. As a consequence of that some method, which enable the quantitative analysis of the influence of entropy and vortex waves to acoustic wave, was suggested.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.17 no.2
• /
• pp.73-81
• /
• 2007

We give two weaknesses of a recently proposed streamcipher MICKEY. We show time-memory-data tradeoff is applicable. We also show that the state update function reduces entropy of the internal state as it is iterated, resulting in keystreams that start out differently but become merged together towards the end.

• Journal of Information Processing Systems
• /
• v.19 no.4
• /
• pp.407-416
• /
• 2023

In order to effectively relieve the traffic pressure of the city, ensure the smooth flow of freight and promote the development of the logistics industry, the selection of appropriate port logistics location is the basis of giving full play to the port logistics function. In order to better realize the selection of port logistics, this paper adopts the entropy weight method to set up a multi-dimensional evaluation index, and constructs the evaluation model of port logistics location. Then through the actual case, from the environmental dimension and economic competition dimension to make choices and analysis. The results show that port d has the largest logistics competitiveness and the highest relative proximity among the three indicators of hinterland city economic activity, hinterland economic structure, and port operation capacity of different port logistics locations, which has absolute advantages. It is hoped that the research results can provide a reference for the multi-dimensional selection of port logistics site selections.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.513-525
• /
• 2024

Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.