• 대한전자공학회논문지SD
• /
• 제45권6호
• /
• pp.16-21
• /
• 2008

본 논문은 새롭게 제안된 단일전자 소자(single-electron device) 및 회로를 이용한 새로운 형태의 홉필드 신경회로망(Hopfield neural network)을 소개한다. 홉필드 신경회로망의 전기적 모델 내부에서 가변저항으로 사용되는 단일전자 시냅스(single-electron synapse)와 비선형 활성함수(nonlinear activation function)로 사용되는 두 단의 단일전자 인버터(single-electron inverter)를 몬테-칼로(Monte-Carlo) 방식의 단일전자 회로 시뮬레이터로 동작을 검증한다.

• 한국지역사회생활과학회지
• /
• 제25권4호
• /
• pp.511-531
• /
• 2014

This study explores supportive and conflictual network structures of elderly single and couple households and analyzes the effects of supportive and conflictual networks on suicidal thoughts by gender and family type. The analysis considered a sample of 522 individuals over the age of 60 who did not live with their adult children. The statistical methods used to analyze data were descriptive statistics, a t-test, a chi-square test and a regression analysis using SPSS WIN 20.0. The results are as follows. First, men and elderly single households had support networks that were smaller than those of women and elderly couple households. The conflictual network of elderly couples households was larger than that of elderly single households. In addition, the larger the network, the more the conflictual was. Second, elderly single households thought about suicide more often than elderly couple households. Third, economic status, the number of adult children, the size of conflictual network and subjective health had considerable influence on suicidal thoughts of elderly single and couple households. The size of the conflictual network had a greater effect on suicidal thoughts of elderly individuals than that of the supportive network. These results have important policy implications for elderly single and couple households.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2000년도 추계학술대회 학술발표 논문집
• /
• pp.471-475
• /
• 2000

Voice and data are transmitted over a single packetized data communications network which is designed for data communications. The public switched telephone network for voice and the packet data network for data are merging into a single data network to get efficiency and to reduce operational cost. However, integrating voice and data transmission over a single data network is not easy because voice should be transmitted without delay but data should be transmitted without error. Advances in technology begin to overcome basic differences. Several integration methods in voice and data will be examined and reviewed here. Moreover, trends and problems on integration will be also discussed.

• 한국멀티미디어학회논문지
• /
• 제23권8호
• /
• pp.1019-1029
• /
• 2020

In this paper, we propose a novel knowledge distillation algorithm to create an compressed deep ensemble network coupled with the combined use of local and global features of face images. In order to transfer the capability of high-level recognition performances of the ensemble deep networks to a single deep network, the probability for class prediction, which is the softmax output of the ensemble network, is used as soft target for training a single deep network. By applying the knowledge distillation algorithm, the local feature informations obtained by training the deep ensemble network using facial subregions of the face image as input are transmitted to a single deep network to create a so-called compressed ensemble DCNN. The experimental results demonstrate that our proposed compressed ensemble deep network can maintain the recognition performance of the complex ensemble deep networks and is superior to the recognition performance of a single deep network. In addition, our proposed method can significantly reduce the storage(memory) space and execution time, compared to the conventional ensemble deep networks developed for face recognition.

• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2009년도 정보통신설비 학술대회
• /
• pp.131-135
• /
• 2009

This paper is the introduction of our work on distributed load scheduling in single-level tree network. In this paper, we derive a new calculation model in single-level tree network and show a closed-form formulation of the time for computation system. There are so many examples of the application of this technology such as distributed database, biology computation on genus, grid computing, numerical computing, video and audio signal processing, etc.

• 한국컴퓨터정보학회논문지
• /
• 제26권6호
• /
• pp.9-17
• /
• 2021

본 논문에서는 기존 네트워크의 물리적인 구조 및 구성을 변경하지 않고 네트워크 보안을 도입할 수 있는 방안으로 "Man In The Middle Attack" 공격 기법을 역이용함으로, Single Ethernet Interface만으로 가상 네트워크 오버레이를 형성하여 논리적인 In-line Mode를 구현하여 외부의 공격으로부터 네트워크를 보호하는 초소형 네트워크 보안 센서와 클라우드서비스를 통한 통합관제 방안을 제안한다. 실험 결과, Single Ethernet Interface만으로 가상 네트워크 오버레이를 형성하여 논리적인 In-line Mode를 구현할 수 있었으며, Network IDS/IPS, Anti-Virus, Network Access Control, Firewall 등을 구현할 수 있었고, 초소형 네트워크보안센서를 클라우드서비스에서 통합관제하는 것이 가능했다. 본 논문의 제안시스템으로 저비용으로 고성능의 네트워크 보안을 기대하는 중소기업에 도움이 되고 또한, IoT 및 Embedded System 분야에 안전·신뢰성을 갖춘 네트워크 보안환경을 제공할 수 있다.

• 한국사회복지학
• /
• 제61권4호
• /
• pp.61-84
• /
• 2009

본 연구는 사회연결망(social network) 접근을 활용하여 한부모 여성의 멘토링 연결망의 특성을 분석하고 멘토링 기능과 임파워먼트에 미치는 효과를 검증하는데 목적을 둔다. 이를 위해 부산 경남의한부모 여성 439명을 대상으로 질문지법을 통해 멘토링 연결망의 특성을 파악하였다. 분석결과 첫째, 한부모 여성들의 멘토링 관계는 연결망 형태임을 확인하였고, 멘토링 연결망의 특성은 대체로 연결정도가 낮고 관계 범위는 협소하며 관계 강도도 약한 편이었다. 둘째, 멘토링 연결망 특성이 멘토링 기능에 미치는 영향을 분석한 결과 경력관련기능은 멘토링 연결망의 특성 중 연결망의 범위와 관계 강도가 유의미한 변수로 나타났고, 심리사회적 기능은 연결망의 크기와 관계 강도가 유의미한 변수였고, 역할모델기능에는 연결망의 크기만이 유의미한 변수로 나타났다. 셋째, 한부모 여성들의 멘토링 연결망의 특성이 임파워먼트에 미치는 직접적인 효과는 적었으나, 멘토링 기능 중 경력관련기능이 임파워먼트에 영향을 미치는 변수로 나타났다. 이러한 연구결과를 바탕으로 제안점과 이론적, 실천적 함의를 제시하였다.

• Journal of Information Processing Systems
• /
• 제18권1호
• /
• pp.12-25
• /
• 2022

In recent years, deep convolutional neural networks have made significant progress in the research of single image super-resolution. However, it is difficult to be applied in practical computing terminals or embedded devices due to a large number of parameters and computational effort. To balance these problems, we propose CSRNet, a lightweight neural network based on channel split residual learning structure, to reconstruct highresolution images from low-resolution images. Lightweight refers to designing a neural network with fewer parameters and a simplified structure for lower memory consumption and faster inference speed. At the same time, it is ensured that the performance of recovering high-resolution images is not degraded. In CSRNet, we reduce the parameters and computation by channel split residual learning. Simultaneously, we propose a double-upsampling network structure to improve the performance of the lightweight super-resolution network and make it easy to train. Finally, we propose a new evaluation metric for the lightweight approaches named 100_FPS. Experiments show that our proposed CSRNet not only speeds up the inference of the neural network and reduces memory consumption, but also performs well on single image super-resolution.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.358-361
• /
• 2021

In order to improve the accuracy of small target detection more effectively, this paper proposes an improved single shot detector (SSD) target detection and recognition method based on cspdarknet53, which introduces lightweight ECA attention mechanism and Feature Pyramid Network (FPN). First, the original SSD backbone network is replaced with cspdarknet53 to enhance the learning ability of the network. Then, a lightweight ECA attention mechanism is added to the basic convolution block to optimize the network. Finally, FPN is used to gradually fuse the multi-scale feature maps used for detection in the SSD from the deep to the shallow layers of the network to improve the positioning accuracy and classification accuracy of the network. Experiments show that the proposed target detection algorithm has better detection accuracy, and it improves the detection accuracy especially for small targets.

• 한국인터넷방송통신학회논문지
• /
• 제10권4호
• /
• pp.151-156
• /
• 2010

본 논문에서는 DVB-H 시스템의 성능측정과 적절한 단일주파수망의 셀 커버리지를 연구하기 위하여 컴퓨터를 이용한 모의실험을 수행하였다. 수행된 결과로부터 2K 전송모드가 8K 전송모드에 비하여 도플러 주파수에 강인함을 확인 할 수 있다. 본 논문의 결과는 단일주파수망 설계에 있어 부분적으로 응용될 수 있다.