• 한국정보통신학회논문지
• /
• 제21권5호
• /
• pp.967-973
• /
• 2017

클러스터링은 균일한 특성을 가지는 데이터를 클러스터로 묶기 위해 사용되는 비교사 학습 방법 중 하나로 다양한 응용에 사용되고 있으며 FCM(Fuzzy C-Means)이 대표적인 방법 중 하나이다. 하지만 FCM에서 주로 사용되는 유클리드 거리 척도는 밀도가 높은 클러스터가 클러스터링 결과에 많은 영향을 미쳐 밀도가 높은 쪽으로 클러스터의 중심을 위치시키는 문제가 있으며, 이를 해결하기 위한 방법 중 하나가 클러스터 중심 사이의 거리가 가능한 멀어지도록 하는 밀도 무관 클러스터링이다. 하지만 밀도 무관 클러스터링 역시 클러스터 중심 사이의 거리를 정확히 제어하기가 어렵다. 이 논문에서는 클러스터 중심 사이의 거리가 멀어지도록 할뿐만이 아니라 클러스터 중심이 밀도가 높은 곳에 위치하도록 하는 항을 추가한 개선된 밀도 무관 클러스터링 방법을 제안한다. 제안하는 방법은 FCM이나 밀도 무관 클러스터링에 비해 실제 클러스터 중심으로 수렴하는 경우가 더 많다는 것을 실험 결과를 통해 확인할 수 있다.

• 대한수학회논문집
• /
• 제20권4호
• /
• pp.813-826
• /
• 2005

Regarding all particles at a fixed site as a cluster, the size of the largest cluster under the zero range invariant measures is well studied by Jeon et al.[5] for the case of density one. Here, the density of the finite zero-range process is given by the ratio between the number m of particles and the number n of sites. In this paper, we study the lower density case, i.e., the case m = o(n). Especially, when m ~ $n^{\beta}$,0 < ${\beta}$ < 1, we show that there is an interesting cutoff point around $\beta$ = 1/2.

• 정보처리학회논문지C
• /
• 제17C권3호
• /
• pp.291-298
• /
• 2010

무선 센서 네트워크에서 센서노드의 수명은 배터리에 의해 제한되므로 에너지는 가장 중요한 고려사항이다. 클러스터링은 네트워크의 에너지 소비를 효율적으로 관리하는데 사용되는 방법 중 하나이며, LEACH는 대표적인 클러스터링 알고리즘이다. LEACH는 센서 노드들의 에너지 소비를 공평하게 분산시키기 위해 에너지 소모적 기능을 하는 클러스터 헤드를 매 라운드마다 무작위로 순환시키는 방법을 사용하고 있다. 클러스터 헤드의 무작위 선정은 매 라운드 최적의 클러스터 헤드 수를 보장해주지 못한다. 그리고 밀도가 높은 클러스터에 위치한 클러스터 헤드는 과부하 상태가 된다. 본 논문에서는 클러스터 헤드의 수를 보장하기 위한 토큰 기반의 클러스터 헤드 선정 알고리즘과 균등한 밀도의 클러스터 형성을 위한 클러스터 선택 알고리즘을 제안한다. 시뮬레이션을 통하여 제안하는 알고리즘이 LEACH 보다 네트워크 수명이 9.3%정도 연장됨을 보여주었다.

• 천문학회보
• /
• 제39권2호
• /
• pp.54-54
• /
• 2014

We use dense redshift surveys of nine galaxy clusters at z ~ 0.2 to compare the galaxy distribution in each system with the projected matter distribution from weak lensing. By combining 2087 new MMT/Hectospec redshifts and the data in the literature, we construct spectroscopic samples within the region of weak-lensing maps of high (70-89%) and uniform completeness. With these dense redshift surveys, we construct galaxy number density maps using several galaxy subsamples. The shape of the main cluster concentration in the weak-lensing maps is similar to the global morphology of the number density maps based on cluster members alone, mainly dominated by red members. We cross correlate the galaxy number density maps with the weak-lensing maps. The cross correlation signal when we include foreground and background galaxies at 0.5zcl < z < 2 zcl is 10 - 23% larger than for cluster members alone at the cluster virial radius. The excess can be as high as 30% depending on the cluster. Cross correlating the galaxy number density and weak-lensing maps suggests that superimposed structures close to the cluster in redshift space contribute more significantly to the excess cross correlation signal than unrelated large-scale structure along the line of sight. Interestingly, the weak-lensing mass profiles are not well constrained for the clusters with the largest cross correlation signal excesses (>20% for A383, A689 and A750). The fractional excess in the cross correlation signal including foreground and background structures could be a useful proxy for assessing the reliability of weak-lensing cluster mass estimates.

• 천문학회지
• /
• 제37권5호
• /
• pp.597-600
• /
• 2004

A substantial number of processes have been suggested as possible contributors to the extragalactic $\gamma$-ray background (EGRB). Yet another contribution to this background will be emission produced in hadronic interactions of cosmic-ray protons with the cluster thermal gas; this class of cosmic rays (CRs) has been shown to be responsible for the EUV emission in the Coma Cluster of galaxies. In this paper we assume the CRs in the Coma Cluster is prototypic of all clusters and derive the contribution to the EGRB from all clusters over time. We examine two different possibilities for the scaling of the CR flux with cluster size: the number density of the CRs scale with the number density of the thermal plasma, and alternatively, the energy density of the CRs scale with the energy density of the plasma. We find that in all scenarios the EGRB produced by this process is sufficiently low that it will not be observable in comparison with other mechanisms that are likely to produce an EGRB.

• Journal of Astronomy and Space Sciences
• /
• 제34권2호
• /
• pp.83-97
• /
• 2017

We present extra-tidal features of spatial configuration of stars around three metal-poor globular clusters (NGC 6266, NGC 6273, NGC 6681) located in the Galactic bulge. The wide-field photometric data were obtained in BVI bands with the MOSAIC II camera at CTIO 4 m Blanco telescope. The derived color-magnitude diagrams (CMDs) contain stars in a total $71^{\prime}{\times}71^{\prime}$ area including a cluster and its surrounding field outside of the tidal radius of the cluster. Applying statistical filtering technique, we minimized the field star contaminations on the obtained cluster CMDs and extracted the cluster members. On the spatial stellar density maps around the target clusters, we found overdensity features beyond the tidal radii of the clusters. We also found that the radial density profiles of the clusters show departures from the best-fit King model for their outer regions which support the overdensity patterns.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2002년도 하계학술대회 논문집
• /
• pp.232-235
• /
• 2002

An 150 kV gas cluster ion accelerator was fabricated and assessed. The change of surface morphology and surface roughness were examined by an atom force microscope (AFM) after irradiation of $CO_2$ gas clusters on Si (100) surfaces at the acceleration voltages of 50 kV. The density of hillocks induced by cluster ion impact was gradually increased with the dosage up to 5$\times$10$^{11}$ ions/$\textrm{cm}^2$. At the boundary of the ion dosage of 10$^{12}$ ions/$\textrm{cm}^2$, the density of the induced hillocks was decreased and RMS (root mean square) surface roughness was not deteriorated further. At the dosage of 5x10$^{13}$ ions/$\textrm{cm}^2$, the induced hillocks completely disappeared and the surface became very flat. In addition, the irradiated region was sputtered. $CO_2$ cluster ions are irradiated at the acceleration voltage of 25 kV to remove hillocks on indium tin oxide (ITO) surface and thus to attain highly smooth surfaces. $CO_2$ monomer ions are also bombarded on the ITO surface at the same acceleration voltage to compare sputtering phenomena. From the AFM results, the irradiation of monomer ions make the hillocks sharper and the surfaces rougher On the other hand, the irradiation of $CO_2$ cluster ions reduces the hight of hillocks and planarize the ITO surfaces. From the experiment of isolated cluster ion impact on the Si surfaces, the induced hillocks m high had the surfaces embossed at the lower ion dosages. The surface roughness was slightly increased with the hillock density and the ion dosage. At higher than a critical ion dosage, the induced hillocks were sputtered and the sputtered particles migrated in order to fill valleys among the hillocks. After prolonged irradiation of cluster ions, the irradiated region was very flat and etched.

• 한국멀티미디어학회논문지
• /
• 제13권6호
• /
• pp.911-923
• /
• 2010

In wireless sensor networks (WSNs), cluster based data routing protocols have the advantages of reducing energy consumption and link maintenance cost. Unfortunately, most of clustering protocols have been designed for uniformly distributed sensor networks. However, some urgent situations do not allow thousands of sensor nodes being deployed uniformly. For example, air vehicles or balloons may take the responsibility for deploying sensor nodes hence leading a normally distributed topology. In order to improve energy efficiency in such sensor networks, in this paper, we propose a new cluster formation algorithm named DAEEC (Density Aware Energy-Efficient Clustering). In this algorithm, we define two kinds of clusters: Low Density (LD) clusters and High Density (HD) clusters. They are determined by the number of nodes participated in one cluster. During the data routing period, the HD clusters help the neighbor LD clusters to forward the sensed data to the central base station. Thus, DAEEC can distribute the energy dissipation evenly among all sensor nodes by considering the deployment density to improve network lifetime and average energy savings. Moreover, because the HD clusters are densely deployed they can work in a manner of our former algorithm EEVAR (Energy Efficient Variable Area Routing Protocol) to save energy. According to the performance analysis result, DAEEC outperforms the conventional data routing schemes in terms of energy consumption and network lifetime.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제17권4호
• /
• pp.267-278
• /
• 2013

We consider zero range processes with density dependent jump rates g given by $g=g(n,k)=g_1(n)g_2(k/n)$ with $g_1(x)=x^{-\alpha}$ and $$g_2(x)=\{^{x^{-\alpha}\;if\;a&lt;x}_{Mx^{-\alpha}\;if\;x{\leq}a}$$. (0.1) In this case, with 1/2 < a < 1 and ${\alpha}$ > 0, we show that non-complete condensation occurs with maximum cluster size an. More precisely, for any ${\epsilon}$ > 0, there exists $M^*$ > 0 such that, for any 0 < M ${\leq}M^*$, the maximum cluster size is between (a - ${\epsilon}$)n and (a + ${\epsilon}$)n for large n. This provides a simple example of non-complete condensation under perturbation of rates which are deep in the range of perfect condensation (e.g. ${\alpha}$ >> 1) and supports the instability of the condensation transition.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2011년도 춘계학술논문집 1부
• /
• pp.306-309
• /
• 2011

A lattice model based molecular clusters is presented to improve a classical equation of state(EOS) for volumetric properties in the critical region. The term is based on the two assumptions: (1) The Helmholtz energy is individually divided into classical and long-range density fluctuation contribution (2) All molecules form cluster near the critical region due to long-range density fluctuation. To formulate such molecular cluster, we extended the Veytsman statistics originally developed for the cluster due to hydrogen bonding. The probability function in the statistics is modified to represent the characteristics of long-range density fluctuation vanishing far from critical region. The proposed fluctuation contribution was incorporated into the Sanchez-Lacombe EOS and the combined model with 6 adjustable parameters has been tested against experimental VLE data for pure compounds. The combined model is found to well represent flatten critical isotherm for methane and top of the coexistence curve for the tested components.