• Proceedings of the Korean Vacuum Society Conference
• /
• 2016.02a
• /
• pp.372.1-372.1
• /
• 2016

Recently, there are a lot of diseases all around the world. Out of them, Atherosclerosis (AS) is the most common cause of stroke, cardiovascular mortality, and myocardial infarction. The macrophage-derived foam cell, which is formed by oxidized low-density lipoprotein (oxLDL), is the crucial marker for AS. In this study, we report a label-free capacitance imaging technique with multi-electrode array (MEA). The lipid-rich aorta arch lesions, which are derived from an apolipoprotein-E receptor-deficient (apoE-/-) mouse, exhibit higher capacitance than the lipid-free aorta arch, allowing the capacitance imaging of lipid region in atherosclerosis. To improve the contacts between MEA and tissue, polypyrrole(PPy)-coated multi walled carbon nanotubes (MWNTs) multi electrode array (PPy-MWNTs-MEA) was fabricated. Compared to TiN-MEA, PPy-MWNTs-MEA yielded lower contact impedance and better capacitance images. In addition, we have also developed a flexible MEA using single walled carbon nanotubes on a PET substrate. The lipid region could be discriminated in the capacitance images of the lipid-rich aorta arch lesions measured using flexible MEA, demonstrating a feasibility of in vivo applications.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.136-138
• /
• 2004

IP망을 활용한 VPN(Virtual Private Network)에서의 QoS 보장 및 광대역 서비스 제공에 대한 해결 방안으로 차세대 광 인터넷을 통한 OVPN(Optical VPN) 기술이 제시되고 있다. 차세대 광 인터넷의 구현이 IP/GMPLS(Generalized Multi-Protocol Label Switching) over DWDM(Dense Wavelength Division Multiplexing) 프로토콜 프레임워크로 표준화되고 있는 현실에 비추어 볼 때 IP/GMPLS over DWDM 백본망을 통한 OVPN은 차세대 가상사설망으로써 멀티미디어 서비스 제공을 위한 최적의 방안이다. 이러한 멀티미디어 서비스 제공을 위한 OVPN에서는 최적의 Optical-LSP(Label Switched Path)의 설립이 요구되고 있다. 따라서 본 논문에서는 차세대 OVPN의 모델 및 망의 blocking probability를 향상하기 위한 미래의 잠재적인 연결 요구에 대해 간섭을 최소화하면서 경로를 설정하는 MW-MIPR 라우팅 알고리즘을 제시하고 이를 라우팅 프로토콜인 OSPF-TE+(OSPF Extensions in Support of Generalized MPLS)에 맵핑하기 위한 방안을 제안한다.

• Journal of Communications and Networks
• /
• v.8 no.4
• /
• pp.480-484
• /
• 2006

This paper briefly reviews ITU-T recommendations associated with multi protocol label switch (MPLS) and Ethernet operation, administration, and maintenance (OAM). The broadband convergence network (BcN) architecture with a centralized network controller is introduced. An aggregation structure employing Ethernet, MPLS, and time division multiplexing (TDM) signal is presented for the BcN. A physical link failure scenario is examined to investigate how the maintenance signal is propagated between different layers.

• Journal of Communications and Networks
• /
• v.9 no.4
• /
• pp.377-382
• /
• 2007

This paper presents a new enhanced dynamic and multilayer protection(DMP) routing scheme that considers cooperation between packet and wavelength switching domain in order to minimize the resource consumption. The paper describes the architecture of the multilayer network scenario and compares the proposed algorithm with other routing mechanisms applying protection at the IP/multi-protocol label switching(MPLS) layer or at the optical layer. Simulation results show that DMP reduces the number of optical-electrical-optical(o-e-o) operations and makes an efficient use of the network resources compared to non-multilayer proposals.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.373-375
• /
• 2023

본 논문에서는 멀티 레이블을 이용한 CNN 구조 활용과 NLP 학습을 이용하여 한국 영화의 장르를 예측하는 방법을 제안한다. 포스터는 영화의 전반적인 내용을 한눈에 알아볼 수 있게 하는 매체이기 때문에 다양한 요소들로 구성되어 있다. 합성곱 신경망(Convolutional neural network)을 활용해, 한국 영화 포스터가 가지는 특징들을 추출하여 영화 장르 분류를 진행하였다. 하지만, 영화의 경우 감독이 생각하는 장르와 관객이 영화를 봤을 때, 느끼는 장르가 다를 수 있다. 그렇기 때문에 장르 예측에 있어서 문제가 발생할 수 있다. 이러한 문제를 완화하기 위해 본 논문에서는 합성곱 신경망 활용뿐만 아니라, 자연어 처리(Natural Language Processing)를 같이 활용한 방법을 제안한다.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1851-1854
• /
• 2002

The Internet with conventional routing scheme cannot meet user demands driven from drastic growth in the Internet user and various service and traffic type. MPLS(Multi Protocol Label Switching) was introduced to the Internet fur solution to resolve this problem. MPLS is a paradigm to integrate higher layer’s software routing functions including layer-3 routing with layer-2 switching. But, the exponential growth of Internet traffic brings out of label space. One scalable solution to cope with this problem is to introduce flow merge technique, i. e. a group of flows is forwarded using the same label. Specially, IETF(Internet Engineering Task Force) recommends that ATM based MPLS system may include VC merge function, so it is scalable to increase of internet traffic. We implemented the MPLS LER system that includes the look-up and forwarding function in incoming path and VC merging function and limited traffic management function in outgoing path. This paper describes the implementation of the LER’s outgoing parts.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.6A
• /
• pp.1060-1066
• /
• 2001

레이어2 스위칭과 레이어3 라우팅의 통합 모델로써 MPLS(Multi-Protocol Label Switching) 환경에서 ATM LSR(Label-Switching Routers)은 백본망에서의 고속 전송이 가능하여 현재의 라우터 구조로써 제안되어지고 있다. MPLS가 코어 라이터로써 적용이 될 경우 확장성을 위해 label merging이라는 기술이 필요하다. VC(Virtual Circuit) merging은 ATM LSR에서 많은 IP 라우터를 하나의 라벨로 매핑을 시키며 수천 개의 목적지에 전송할 수 있는 확장성 있는 매핑 기술이다. VC merging은 같은 목적지인 다른 패킷들 간의 셀들의 섞임을 방지하기 위해 재 조합 버퍼가 요구된다. 재 조합 버퍼 사용시 일시적인 체증 현상이 발생하며 Non-VC merging과 비교시 많은 셀 손실과 많은 버퍼를 요구하게 된다. 본 논문에서는 RED(Random Early Detection) 알고리즘을 적용하여 VC merging이 필요한 버퍼의 요구량과 셀 손실을 줄였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.54-57
• /
• 2007

Much researches have studied for seamless mobility service. Those focused on minimizing the delay time due to the handover. In this paper, we suggest seamless mobility service with the network-based preemptive operations. With these operations, if it's found that the MT(Mobile Terminal)'s handover using L2-trigger event, old access network buffers the delivering data. Therefore this can decrease the data drop rates. And also, this can deal with the ping-pong's phenomenon of MT. At the end of MT's movement, these operations can provide seamless mobility service sending buffered data after checking the MT's movement. This mechanism uses MPLS-LSP(MultiProtocol Label Switching-Label Switched Path) in core network for fast process.

• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.17-25
• /
• 2021

When a person wears clothing, it is important to configure an outfit appropriate to the intended occasion. Therefore, T.P.O(Time, Place, Occasion) of the outfit is considered in various fashion recommendation systems based on artificial intelligence. However, there are few studies that directly infer the T.P.O from outfit images, as the nature of the problem causes multi-label and class imbalance problems, which makes model training challenging. Therefore, in this study, we propose a model that can infer the T.P.O of outfit images by employing a label-distribution-aware margin(LDAM) loss function. Datasets for the model training and evaluation were collected from fashion shopping malls. As a result of measuring performance, it was confirmed that the proposed model showed balanced performance in all T.P.O classes compared to baselines.