• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.5C
• /
• pp.403-412
• /
• 2008

This paper presents the methods which obtain the solution of Wiener-Hopf equation by LMS algorithm and get the coefficient of TDL filter in lattice filter directly. For this result, we apply an orthogonal input signal generated by lattice filter into an equivalent Wiener-Hopf equation and shows the scheme that can obtain the solution by using the MMSE algorithm. Conventionally, the method like aforementioned scheme can get an error and regression coefficient recursively. However, in this paper, we can obtain an error and the coefficients of TDL filter recursively. And, we make an theoretical analysis on the convergence characteristics of the proposed algorithm. Then we can see that the result is similar to conventional analysis. Also, by computer simulation, we can make sure that the proposed algorithm has an excellent performance.

• Journal of KIISE:Databases
• /
• v.31 no.6
• /
• pp.710-720
• /
• 2004

As XML documents have much richer information than plain texts, we can perform very elaborated, fine-grained search which was difficult in past years. However, as the cost of finer grained element level search is very high, the processing overhead has become a new challenge. We propose an inverted index structure called PIX, which reduces the number of elements processed by partitioning elements according to their match potentiality. We choose a base level and partition elements according to whether they have possibility of having a common ancestor higher than the level. We also propose partition merging technique by which we can get same results as unpartitioned case. Our experimental results show that the index partitioning strategy can reduce processing time considerably.

• Journal of KIISE:Information Networking
• /
• v.30 no.5
• /
• pp.641-648
• /
• 2003

In this paper, we propose a general analysis tool that derives guaranteed delay bounds for real-time sessions when those sessions pass through heterogeneous schedulers in general packet networks. This tool characterizes each different scheduler by a homogeneous service curve server. We show that service curve servers can characterize a strictly larger class of schedulers than LR servers. That is, we can characterize not only all LR servers but also RC-EDF and SC scheduler by service curve servers. Even with such advantage over LR servers, service curve servers provide accurate analysis results. We prove mathematically that if we analyze a network that can be analyzed by both LR servers and service curve servers,,both cases yield the same delay bound.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.1
• /
• pp.165-174
• /
• 2019

They are frequently used today in public places such as street, subway, school or library. Since users can sometimes print documents that contain confidential data using Printer Kiosks, the devices should store and manage the documents securely. In this paper, we identify potential security threats in Printer Kiosks and suggest practical attack scenarios that can take place. To show the feasibility of suggested attack, we analyzed network traffic that were generated by the real Printer Kiosk device. As a result of our analysis, we have found that attackers can access other users' scanned files and access other users' documents from Printer Kiosk's home page. We confirmed that using our attack, we could retrieve other users' personal data.

• Proceedings of the Korean Society of Computational and Applied Mathematics Conference
• /
• 2003.09a
• /
• pp.17.1-17
• /
• 2003

We study averaging properties in Banach spaces, First, we seek an averaging property equivalent to the reflexivity in Banach spaces. Second, we investigate averaging properties in Banach space using the Brunet-Sucheston's spreading model.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.13-15
• /
• 2006

When we use multilayer perceptron model for pattern classification probmems, we allocate one output node for each class. In this paper, we increase the number of output nodes for each class and investigate the performance of multilayer perceptrons through the simulation of isolated-word recognition problems.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.15 no.3
• /
• pp.198-207
• /
• 2012

In this study, we investigated the effect of ocean fertilization by deep sea water, using an ecosystem model which contains not only phytoplankton but also zooplankton. The model is based on NEMURO which consists of eleven compartments - two species of phytoplankton, three species of zooplankton, $NO_3$, $NH_4$, $Si(OH)_4$, particulate organic nitrogen, dissolved organic nitrogen and particulate silicon. We introduced nitrogen cell quota in the both species of phytoplankton, and silicon cell quota in the large phytoplankton in addition to the eleven compartments of NEMURO. We made the experiment at Izu Oshima Island in order to investigate the effect of ocean fertilization. In this experiment, we could not find clear differences between the cases with and without deep sea water. We investigated the causes of the experiment results by the model simulations. One of the causes was high concentrations of nutrients in surface seawater used in the experiment. Another was that the increase of total concentration of inorganic nitrogen does not necessarily accelerate the photosynthetic rate because inorganic nitrogen uptake rate is related to the ratio of $NO_3$ to $NH_4$. Because the model can represent the results of the experiment, we investigated the effect of ocean fertilization by deep sea water using this model. We found that the effect of ocean fertilization hardly appeared when the interval of the addition of deep sea water was too short, or the amount of deep sea water was too much. It is supposed that if the addition of deep sea water is too frequent or too much, the dilution of plankton's concentrations will exceed the effect of promoting phytoplankton's photosynthesis.

• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.929-938
• /
• 2004

This paper proposes a pose-invariant face recognition method using cylindrical model and stereo camera. We divided this paper into two parts. One is single input image case, the other is stereo input image case. In single input image case, we normalized a face's yaw pose using cylindrical model, and in stereo input image case, we normalized a face's pitch pose using cylindrical model with previously estimated pitch pose angle by the stereo geometry. Also, since we have an advantage that we can utilize two images acquired at the same time, we can increase overall recognition performance by decision-level fusion. Through representative experiments, we achieved an increased recognition rate from 61.43% to 94.76% by the yaw pose transform, and the recognition rate with the proposed method achieves as good as that of the more complicated 3D face model. Also, by using stereo camera system we achieved an increased recognition rate 5.24% more for the case of upper face pose, and 3.34% more by decision-level fusion.

• Journal of the Korean School Mathematics Society
• /
• v.6 no.2
• /
• pp.71-85
• /
• 2003

In this paper we study the usefulness of Maple in school calculus education. The use of computer and calculator is debated in many aspect in mathematics education. By the computer visualization of mathematical image and proper use of computer we can teach inductively and intuitively the mathematical concept and give rise to the students' interest. Maple is very popular in college but is not in middle school because of language. Maple application which is made by Maplet is very useful multimedia teaching-learning tools. We introduce the use of Maplet and some application of the calculus course which we made.

• Journal of Energy Engineering
• /
• v.13 no.2
• /
• pp.128-132
• /
• 2004

This paper presents a new methodology to get pareto-optimal solution for generation planning. First, we apply dynamic programming, and we can get an optimal economic dispatch considering total quantity of contamination for the specified term. Second, we developed a method which can get pareto-optimal solution. This solution is consisted of a set of optimal generation planning. As a result, decision maker can get pareto-optimal solutions, and can choose a solution. We applied this method to the test system, and showed the usefulness.