• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2001년도 하계학술대회 논문집
• /
• pp.636-639
• /
• 2001

InGaN/GaN multiple quantum wells (MQWs) grown with various growth interruptions between the InGaN well and GaN barrier by metal-organic chemical vapor deposition were investigated using photoluminescence, high-resolution transmission electron microscopy, and energy filtered transmission electron microscopy (EFTEM). The luminescence intensity of the MQWs with growth interruptions is abruptly reduced compared to that of the MQW without growth interruption. Also, as the interruption time increases the peak emission shows a continuous blue shift. Evidence of indium clustering is directly observed both by using an indium ratio map of the MQWs and from indium composition measurements along an InGaN well using EFTEM. The higher intensity and lower energy emission of light from the MQW grown without interruption showing indium clustering is believed to be caused by the recombination of excitons localized in indium clustering regions and the increased indium composition in these recombination centers.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2004년도 가을 학술발표논문집 Vol.31 No.2 (1)
• /
• pp.697-699
• /
• 2004

Despite many successful spectral clustering algorithm (based on the spectral decomposition of Laplacian(1) or stochastic matrix(2) ) there are several unsolved problems. Most spectral clustering Problems are based on the normalized of algorithm(3) . are close to the classical graph paritioning problem which is NP-hard problem. To get good solution in polynomial time. it needs to establish its convex form by using relaxation. In this paper, we apply a novel optimization technique. semidefinite programming(SDP). to the unsupervised clustering Problem. and present a new multiple Partitioning method. Experimental results confirm that the Proposed method improves the clustering performance. especially in the Problem of being mixed with non-compact clusters compared to the previous multiple spectral clustering methods.

• 스마트미디어저널
• /
• 제8권1호
• /
• pp.43-50
• /
• 2019

수많은 디바이스들이 무선 네트워크를 통해 연결 되고 있고, 이러한 연결을 효율적으로하기 위한 연구들이 진행되고 있다. 많은 연구에서 효율적인 디바이스 관리를 위해 클러스터링을 사용하고 있지만 클러스터의 특정 노드에 부하가 집중되는 경우가 많아 전체 네트워크가 불안정해질 수 있다. 이러한 문제를 해결하기 위해 본 논문에서는 센서 노드의 효율적인 관리를 위해 중앙 집중식 불균등 체인 클러스터 스케줄링 모델을 제안한다. 클러스터의 구성을 위해 클러스터 헤드 범위와 기지국까지의 거리를 기반으로 하고, 기지국의 위치가 동일하지 않은 동심 체인 클러스터링을 구축하기 위해 주벡터 투사 기법을 사용한다. 데이터의 전송은 다중 무선 액세스 인터페이스인 MIMO(Multiple-Input Multiple-Output)를 활용한다. 실험을 통해 클러스터 헤드의 에너지 소비를 줄이고 네트워크 수명이 향상됨을 보인다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권4호
• /
• pp.254-268
• /
• 2013

For real-world clustering tasks, the input data is typically not easily separable due to the highly complex data structure or when clusters vary in size, density and shape. Kernel-based clustering has proven to be an effective approach to partition such data. In this paper, we provide an overview of several fuzzy kernel clustering algorithms. We focus on methods that optimize an fuzzy C-mean-type objective function. We highlight the advantages and disadvantages of each method. In addition to the completely unsupervised algorithms, we also provide an overview of some semi-supervised fuzzy kernel clustering algorithms. These algorithms use partial supervision information to guide the optimization process and avoid local minima. We also provide an overview of the different approaches that have been used to extend kernel clustering to handle very large data sets.

• 한국인터넷방송통신학회논문지
• /
• 제10권6호
• /
• pp.145-152
• /
• 2010

본 연구에서는 시계열 자료를 베이지안 정보기준을 통해 클러스터링 한다. 보다 안정적인 클러스터를 생산하기 위해 다중 중심점을 모델링한 후 이를 이용하여 클러스터를 생성시킨다. 대상 시계열 자료에 대해 예측할 경우 클러스터에 속한 시계열 자료 중 가장 유사한 시계열 자료를 선택하여 모델링한다. 모델로부터 마코프 규칙을 유도해 내고 이 규칙을 이용해 예측정확도를 측정한다. 시계열 자료를 단독으로 모델링한 후 예측한 결과보다 클러스터에 속한 유사시계열 모델링을 통한 예측정확도가 좀 더 높았음을 확인하였다.

• 한국생물공학회:학술대회논문집
• /
• 한국생물공학회 2003년도 생물공학의 동향(XIII)
• /
• pp.746-749
• /
• 2003

Alignment of 2D-gel images of biological samples can visualize the difference of expression profiles and also inform us candidates of protein spots to be further analyzed. However, comparison of two proteome images between case and control does not always successfully identify differentially expressed proteins due to sample-to-sample variation. Because of poor reproducibility of 2D-gel electrophoresis, sample-by-sample variations and inconsistent electrophoresis conditions, multiple number of 2D-gel image must be processed to align each other to visualize the difference of expression profiles and to deduce the protein spots differentially expressed with reliability. Alignment of multiple 2D-Gel images and their clustering were carried out by applying various algorithms and statistical methods. In order to align multiple images, multiresolution-multilevel algorithm was found out to be suitable for fast alignment and for distorted images. Clustering of 12 different images implementing a k-means algorithm gives a phylogenetic tree of distance map of the proteomes. Microsoft Visual C++ was used to implement the algorithms in this work.

• 지능정보연구
• /
• 제6권1호
• /
• pp.97-112
• /
• 2000

The basic idea of case-based reasoning is to solve a new problem using the previous problem-solving experiences. In this research we develop a case-based reasoning system for equipment malfunction diagnosis. We first divide the case base into clusters using the case-based clustering technique. Then we develop an appropriate case-based diagnostic system for each cluster. In other words for individual cluster a different case-based diagnostic system which uses different weights for attributes is developed. As a result multiple case-based reasoning system are operating to solve a diagnostic problem. In comparison to the performance of the single case-based reasoning system our system reduces the computation time by 50% and increases the accuracy by 5% point.

• IEIE Transactions on Smart Processing and Computing
• /
• 제5권2호
• /
• pp.85-93
• /
• 2016

The existence of correlation characteristics brings significant potential advantages to the development of efficient routing protocols in wireless sensor networks. This research proposes a new simple method of clustering sensor nodes into correlation groups in multiple-correlation areas. At first, the evaluation of joint entropy for multiple-sensed data is considered. Based on the evaluation, the definition of correlation region, based on entropy theory, is proposed. Following that, a correlation clustering scheme with less computation is developed. The results are validated with a real data set.

• 한국정밀공학회지
• /
• 제16권5호통권98호
• /
• pp.48-57
• /
• 1999

Applications of thresholding technique are based on the assumption that object and background pixels in a digital image can be distinguished by their gray level values. For the segmentation of more complex images, it is necessary to resort to multiple threshold selection techniques. This paper describes a new method for multiple threshold selection of gray level images which are not clearly distinguishable from the background. The proposed method consists of three main stages. In the first stage, a probability distribution function for a gray level histogram of an image is derived. Cluster points are defined according to the probability distribution function. In the second stage, fuzzy partition matrix of the probability distribution function is generated through the fuzzy clustering process. Finally, elements of the fuzzy partition matrix are classified as clusters according to gray level values by using max-membership method. Boundary values of classified clusters are selected as multiple threshold. In order to verify the performance of the developed algorithm, automatic inspection process of ball grid array is presented.

• ETRI Journal
• /
• 제43권2호
• /
• pp.221-231
• /
• 2021

To resolve energy depletion issues in massive Internet of Things sensor networks, we developed a set of distributed energy beamforming methods with one-bit feedback and clustering for multi-node wireless energy transfer, where multiple singleantenna distributed energy transmitters (Txs) transfer their energy to multiple nodes wirelessly. Unlike previous works focusing on distributed information beamforming using a single energy receiver (Rx) node, we developed a distributed energy beamforming method for multiple Rx nodes. Additionally, we propose two clustering methods in which each Tx node chooses a suitable Rx node. Furthermore, we propose a fast distributed beamforming method based on Tx sub-clustering. Through computer simulations, we demonstrate that the proposed distributed beamforming method makes it possible to transfer wireless energy to massive numbers of sensors effectively and rapidly with small implementation complexity. We also analyze the energy harvesting outage probability of the proposed beamforming method, which provides insights into the design of wireless energy transfer networks with distributed beamforming.