• 한국CDE학회논문집
• /
• 제10권6호
• /
• pp.421-431
• /
• 2005

This paper introduces a new approach for saving build time of hybrid rapid prototyping by decomposing a part into minimum number of layers. In the hybrid rapid prototyping, a part of a complicated shape is realized by adding layers of a simpler shape, each of which is obtained by machining a sheet of constant thickness from its top and bottom surfaces. Thus it is desired to decompose a given part into the minimum number of layers while guaranteeing each layer to be fabricated from the given sheets using a 3-axis milling machine. To satisfy these requirements, a concave edge-based algorithm is proposed to decompose a part into layers by considering the tool accessibility, the total number of layers, and the allowable sheet thickness.

• 정보와 통신
• /
• 제24권2호
• /
• pp.64-73
• /
• 2007

Recently, cross-layer design approach has been greatly attracting researchers' attention as an alternative for improving the performance of wireless data networks. The main reason why cross-layer approaches are particularly well suited for wireless networks is that there exists direct coupling between physical layer and upper layers. Therefore, with cross-layer approach, the protocol designers try to exploit the interaction between layers and promote adaptability at all layers, based on information exchange between layers. In this article we focus on the cross-layer engineering for high data-rate mobile Internet services through cellular networks. First, the general considerations in cross-layer engineering are outlined. Then, we discuss the common approach in literatures, which mainly deals with adaptability in physical and medium access control layer. Finally, we show that the cross-layer engineering taking account of all layers is more adequate for the mobile Internet services cellular network.

• Journal of Magnetics
• /
• 제14권3호
• /
• pp.104-107
• /
• 2009

The effects of the magnetic property variation on current-induced magnetization switching in magnetic tunnel junction with perpendicular magnetic anistoropy (PMA) and the soft magnetic polarization-enhancement layers (PELs) inserted between the layers with PMA and the MgO layer was studied. A micromatnetic model was used to estimate the switching time of the free layer by different applied current densities, with changing saturation magnetization ($M_s$) of the PELs, interlayer exchange coupling between PMA layers and PELs. The switching time could be significantly reduced at low current densities, by increasing $M_s$ of PELs and decreasing interlayer exchange coupling.

• ETRI Journal
• /
• 제35권4호
• /
• pp.730-733
• /
• 2013

In this work, buffer layers with various conditions are inserted at an n/i interface in hydrogenated amorphous silicon semitransparent solar cells. It is observed that the performance of a solar cell strongly depends on the arrangement and thickness of the buffer layer. When arranging buffer layers with various bandgaps in ascending order from the intrinsic layer to the n layer, a relatively high open circuit voltage and short circuit current are observed. In addition, the fill factors are improved, owing to an enhanced shunt resistance under every instance of the introduced n/i buffer layers. Among the various conditions during the arrangement of the buffer layers, a reverse V shape of the energy bandgap is found to be the most effective for high efficiency, which also exhibits intermediate transmittance among all samples. This is an inspiring result, enabling an independent control of the conversion efficiency and transmittance.

• 한국컴퓨터정보학회논문지
• /
• 제21권1호
• /
• pp.1-8
• /
• 2016

Convolutional neural network (CNN) consists of a few pairs of both convolution layer and subsampling layer. Thus it has more hidden layers than multi-layer perceptron. With the increased layers, the size of convolution mask ultimately determines the total number of weights in CNN because the mask is shared among input images. It also is an important learning factor which makes or breaks CNN's learning. Therefore, this paper proposes the best method to choose the convolution size and the number of layers for learning CNN successfully. Through our face recognition with vast learning examples, we found that the best size of convolution mask is 5 by 5 and 7 by 7, regardless of the number of layers. In addition, the CNN with two pairs of both convolution and subsampling layer is found to make the best performance as if the multi-layer perceptron having two hidden layers does.

• 한국표면공학회:학술대회논문집
• /
• 한국표면공학회 2003년도 춘계학술발표회 초록집
• /
• pp.61-61
• /
• 2003

It is important to know SIze distribution of defects In electroplated trivalent chrome layers because it significantly influences on performance of the layers. Most of the nano-scale defects are able to be introduced by hydrogen evolution during the plating. Little information is available on the nano-size defects. In this study, SANS was applied to determine the size distribution of nano-scale defects in the trivalent chrome layers prepared in a formate bath. The defect size and distribution was dependent upon plating conditions such as current density and applied voltage. SANS is one of useful techniques to determine the nano-scale defect in the electroplated layers.

• Bulletin of the Korean Chemical Society
• /
• 제18권11호
• /
• pp.1195-1199
• /
• 1997

Anthraquinone-2-sulfonate (AQS) release from poly(N-methylpyrrole anthraquinone-2-sulfonate) (PNMP-AQS) was investigated at open circuit and compared with electrochemically stimulated release during potential cycling. It was found that the fast AQS release from PNMP-AQS single layers is substantially retarded and the amounts of spontaneously and electrochemically releasable AQS can be reduced by constructing bilayers, consisting of PNMP-AQS inner layers and PNMP outer layers. PNMP-Cl outer layers exhibited higher effectiveness for entrapping AQS within inner layers than PNMP/poly(styrene slfonate). The effects of outer layer thicknesses on AQS release were also examined with PNMP-AQS:PMP-Cl. The electroactivity enhancement of PNMP-AQS:PNMP-Cl bilayers due to entrapped AQS was confirmed by chronocoulometry.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1993년도 학술논문발표회 논문집 제12권 1호
• /
• pp.190-194
• /
• 1993

In the ocean bubble layers play a significant role in the sound propagation as well as sound generation. Most of early works dealt with acoustic properties of the bubble layers with sharp flat boundaries. Bubble layers with sharp flat boundaries. Bubble layers with transition sublayers are more likely in the coean. In this paper a theory of sound propagation through plane bubble layers with transition sublayers at both borders was developed. It shows that the reflection and transmission coefficients depend on the thickness of transition sublayers. The theory with thicker transition sublayers shows weaker resonance properties of bubble layer. It gives better presentation for the peculiar behavior of the experimental data than that with sharp flat boundaries.

• 한국주조공학회지
• /
• 제25권4호
• /
• pp.168-172
• /
• 2005

In this study, the effects of chill layers occurred in shot sleeve on the molten metal filling were analyzed through computer simulation and the behavior of chill layers with temperature variation of shot sleeve set from 200 to $280^{\circ}C$ was also investigated. The simulation results showed the chill layers set in the in-gates during the injection process change the main filling direction and cause turbulent flow pattern, resulting in porosities inside the castings. The amount of chill layers with the increasing temperature of shot sleeve was considerably reduced. And particularly, at the setting temperature of $280^{\circ}C$ by heat control unit, the big reduction in chill layers, excellent trimmed surface and the highest densification were achieved, suggesting that as the optimal sleeve condition in diecasting, especially for the highly complex parts like valve body.

• 한국멀티미디어학회논문지
• /
• 제23권7호
• /
• pp.803-811
• /
• 2020

This paper proposes an image classification algorithm that transforms the number of convolution layers in the residual block of ResNet, CNN's representative method. The proposed method modified the structure of 34/50 layer of ResNet structure. First, we analyzed the performance of small and many convolution layers for the structure consisting of only shortcut and 3 × 3 convolution layers for 34 and 50 layers. And then the performance was analyzed in the case of small and many cases of convolutional layers for the bottleneck structure of 50 layers. By applying the results, the best classification method in the residual block was applied to construct a 34-layer simple structure and a 50-layer bottleneck image classification model. To evaluate the performance of the proposed image classification model, the results were analyzed by applying to the cifar10 dataset. The proposed 34-layer simple structure and 50-layer bottleneck showed improved performance over the ResNet-110 and Densnet-40 models.