• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.1
• /
• pp.94-99
• /
• 2012

This paper discusses a decentralized sampled-data control problem for large-scale nonlinear systems. The system is represented in Takagi-Sugeno's form. Next, we design a decentralized analog controller based on the overlapping decomposition technique. The final step is to apply the intelligent digital redesign scheme for converting the analog controller into the sampled-data one. Design condition is represented in terms of linear matrix inequalities. A simulation result is provided for the effectiveness of the proposed design method.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.157-161
• /
• 2016

본 논문에서는 클러스터 간의 중복을 허용한 계층적 클러스터링(hierarchical clustering) 기법에 적합한 클러스터 간 유사도 평가방법(linkage metric)을 제안하였다. 클러스터 간 유사도 평가방법은 계층적 클러스터링에서 클러스터를 통합하거나 분해하는데 쓰이며 사용된 방법에 따라 클러스터링의 결과가 다르게 형성된다. 기존의 클러스터 간 유사도 평가방법인 single linkage, complete linkage, average linkage 중 single linkage와 complete linkage는 클러스터 간 중복이 허용된 환경에서 정확도가 낮은 문제점이 있고, average linkage는 정확도가 두 방법에 비해 높지만 계산 시간 소요가 크다는 단점이 있다. 따라서 본 논문에서는 기존의 average linkage를 개선하여 중복된 데이터에 의한 필요 계산량을 크게 줄임으로써 시간적 성능이 우수한 클러스터 간 유사도 평가방법을 제안하였다. 또한, 제안된 방법을 기존 방법들과 비교실험하여 중복을 허용하는 계층적 클러스터링 환경에서 정확도는 비슷하거나 더 높고, average linkage에 비해 계산량이 감소됨을 확인하였다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.157-161
• /
• 2016

본 논문에서는 클러스터 간의 중복을 허용한 계층적 클러스터링(hierarchical clustering) 기법에 적합한 클러스터 간 유사도 평가방법(linkage metric)을 제안하였다. 클러스터 간 유사도 평가방법은 계층적 클러스터링에서 클러스터를 통합하거나 분해하는데 쓰이며 사용된 방법에 따라 클러스터링의 결과가 다르게 형성된다. 기존의 클러스터 간 유사도 평가방법인 single linkage, complete linkage, average linkage 중 single linkage와 complete linkage는 클러스터 간 중복이 허용된 환경에서 정확도가 낮은 문제점이 있고, average linkage는 정확도가 두 방법에 비해 높지만 계산 시간 소요가 크다는 단점이 있다. 따라서 본 논문에서는 기존의 average linkage를 개선하여 중복된 데이터에 의한 필요 계산량을 크게 줄임으로써 시간적 성능이 우수한 클러스터 간 유사도 평가방법을 제안하였다. 또한, 제안된 방법을 기존 방법들과 비교 실험하여 중복을 허용하는 계층적 클러스터링 환경에서 정확도는 비슷하거나 더 높고, average linkage에 비해 계산량이 감소됨을 확인하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.3
• /
• pp.234-239
• /
• 2015

Non-Negative Matrix Factorization (NMF) is a method for updating dictionary and gain in alternating manner. Due to ease of implementation and intuitive interpretation, NMF is widely used to detect and separate overlapping sound events. However, NMF that utilizes non-negativity constraints generates parts-based representation and this distinct property leads to a dictionary containing fragmented acoustic events. As a result, the presence of shared basis results in performance degradation in both separation and detection tasks of overlapping sound events. In this paper, we propose a new method that utilizes K-Singular Value Decomposition (K-SVD) based dictionary to address and mitigate the part-based representation issue during the dictionary learning step. Subsequently, we calculate the gain using NMF in sound event detection step. We evaluate and confirm that overlapping sound event detection performance of the proposed method is better than the conventional method that utilizes NMF based dictionary.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.4
• /
• pp.364-370
• /
• 2004

In this paper, we propose a FDI method, which comes from singular value decomposition of measurement matrix fur redundant sensors. We analyze the performance of the proposed FDI method by comparing with the GLT method in two ways such as FDI performance and GN＆C performance. Also, we propose a GN＆C performance index by combining FDI and GN＆C performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.8C
• /
• pp.703-711
• /
• 2007

In this paper, we propose a multi-view coding(MVC) algorithm with considering view scalability. The proposed algorithm has a high compression efficiency by reducing inter-view redundancy through inter-view decomposition, and adaptively reconstructs a multi-view video from an encoded bit stream. Furthermore, a reference view can be decoded by a traditional H.264/AVC, and the other views are adaptively decoded at the receiver by filtering to support view scalability. Experimental results show that the proposed algorithm performed better than the conventional H.264 codec even though it offers the view scalability.

• Computational Structural Engineering
• /
• v.11 no.1
• /
• pp.205-216
• /
• 1998

A solution method is presented to solve the eigenvalue problem arising in the dynamic analysis of nonclassicary damped structural systems with multiple eigenvalues. The proposed method is obtained by applying the modified Newton-Raphson technique and the orthonormal condition of the eigenvectors to the linear eigenproblem through matrix augmentation of the quadratic eigenvalue problem. In the iteration methods such as the inverse iteration method and the subspace iteration method, singularity may be occurred during the factorizing process when the shift value is close to an eigenvalue of the system. However, even though the shift value is an eigenvalue of the system, the proposed method provides nonsingularity, and that is analytically proved. Since the modified Newton-Raphson technique is adopted to the proposed method, initial values are need. Because the Lanczos method effectively produces better initial values than other methods, the results of the Lanczos method are taken as the initial values of the proposed method. Two numerical examples are presented to demonstrate the effectiveness of the proposed method and the results are compared with those of the well-known subspace iteration method and the Lanczos method.

• Annual Conference of KIPS
• /
• 2006.05a
• /
• pp.199-202
• /
• 2006

실시간 시스템에서는 정해진 시간 안에 작업을 수행해야 하는 것이 가장 중요하다. 때문에 실시간 시스템에서의 응답시간 위반은 물질 또는 인명 피해와 직결된다. 이에 응답시간을 보장하기 위해 실시간 시스템을 분석하는 기법들에 대한 많은 연구가 진행되었다. 그러나 기존의 분석 방법들은 최악실행시간을 도출하기위해 실시간 프로그램의 흐름을 분석하거나 분석을 위한 제약을 생성할 때 부하가 생기는 문제점을 가지고 있다. 이러한 문제를 해결하기 위하여 본 논문에서는 프로그램에서 나타나는 함수 또는 변수 등이 중복해서 사용되는 특성들을 이용하여 실행시간 분석에 대한 부하를 줄일 수 있는 방법을 제안하고, 제안한 방법을 기반으로 실시간 프로그램에서 실행시간을 예측할 때 필요한 기본 자료들을 도출할 수 있는 소스코드 분석 도구를 제안한다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.7
• /
• pp.1014-1018
• /
• 2012

This paper presents a decentralized controller design for formation flying of multiple satellites based on the overlapping decomposition technique. Each satellite is assumed to avail only the information of its own and in front of itself, which restricts the structure of a controller gain matrix to an overlapped form. The concerned large-scale system is expanded using the overlapping decomposition technique. Design condition is represented in terms of linear matrix inequalities with small-scale systems in a decentralized form, based on the expanded system. The resulting controller is contracted to the original overlapped form so as to close the original system. A numerical simulation shows the effectiveness of the proposed technique.

• The KIPS Transactions:PartD
• /
• v.12D no.5 s.101
• /
• pp.663-672
• /
• 2005

This Paper Proposes an approach that removes the redundancy of Path information and uses an inverted index, as an efficient way to store a large volume of XML documents and to retrieve wanted information from there. An XML document is decomposed into nodes based on its tree structure, and stored in relational tables according to the node type, with path information from the root to each node. The existing methods using path information store data for all element paths, which cause retrieval performance to be decreased with increased data volume. Our approach stores only data for leaf element path excluding internal element paths. As the inverted index is made by the leaf element path only, the number of posting lists by key words become smaller than those of the existing methods. For the storage and retrieval of U data, our approach doesn't require the XML schema information of XML documents and any extension of relational database. We demonstrate the better performance of on approach than the existing approaches within the scope of our experiment.