• 한국멀티미디어학회논문지
• /
• 제14권10호
• /
• pp.1303-1310
• /
• 2011

본 논문은 인프라 클라우딩 환경에서 데이터를 소유한 소유자 노드와 데이터를 보관하는 보관 노드 그리고 데이터를 검증하는 검증 노드로 구성된 자가 조직 저장 매체 보안을 위한 3자간 협상 프로토콜을 설계를 제안한다. 제안한 자가 조직 저장 매체의 보안기법은 보관 노드의 데이터 검증을 검증 노드에게 위임함으로써 데이터 검증의 효율성을 증가시키고, EC-DH 알고리즘을 이용하여 생성된 암호키와 저장 매체 내의 인증서로 보안을 강화시켰다. 또한, 자가 조직 저장 매체를 구성할 때, 3자간 인증키를 설정하여 외부적인 플러딩 공격 방지하고, 검증노드의 개수를 제한함으로써 내부적인 플러딩 공격을 방지하였다. 그리고 검증단계에서 발생할 수 있는 재전송 공격은 검증을 요청할 때마다 새롭게 생성된 Seed 값을 이용하여 자동적으로 재전송 공격을 탐지하도록 하였다.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2009년 추계학술발표대회
• /
• pp.73-73
• /
• 2009

In this sutdy, a new class ACA(Anisotropic Conductive Adhesive) with low-melting-point alloy(LMPA) and self-organized interconnection method were developed. This developed self-organized interconnection method are achieved by the flow, melting, coalescence and wetting characteristics of the LMPA fillers in ACA. In order to observe self-interconnection characteristic, the QFP($14{\times}14{\times}2.7mm$ size and 1mm lead pitch) was used. Thermal characteristic of the ACA and temperature-dependant viscosity characteristics of the polymer were observed by differential scanning calorimetry(DSC) and torsional parallel rheometer, respectively. A electrical and mechanical characteristics of QFP bonding were measured using multimeter and pull tester, respectively. Wetting and coalescence characteristics of LMPA filler particles and morphology of conduction path were observed by microfocus X-ray inspection systems and cross-sectional optical microscope. As a result, the developed self-organized interconnection method has a good electrical characteristic($2.41m{\Omega}$) and bonding strength(17.19N) by metallurgical interconnection of molten solder particles in ACA.

• 정보처리학회논문지B
• /
• 제8B권6호
• /
• pp.651-656
• /
• 2001

본 논문은 기존의 신경망을 이용한 세선화 방법 중에서 자기 구성 그래프(Self-Organized Graph：SOG) 세선화 기법의 우수한 세선화 결과를 유지하면서, 수행 속도를 향상시키기 위하여 Kohonen Features Map의 새로운 점증 기법을 변형된 SOG에 적용한 개선된 SOG(Improved SOG：$SOG^*$) 세선화 기법을 제안한다. 실험 결과로써 숫자와 문자 모두 기존의 SOG와 같은 우수한 세선화 결과를 나타내며, O((logM)3)의 시간 복잡도를 가지는 속도 향상을 이루었다. 따라서 본 논문에서 제안한 방법은 숫자 또는 문자 인식에 있어 특징 추출의 빠른 전처리 과정으로 사용할 수 있다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.96.3-96
• /
• 2001

In this paper the self-organized and fuzzy inference used stochastic gradient ascent method is proposed. Fuzzy rule and fuzzy set increase as occasion demands autonomously according to the observation information. And two rules(or two fuzzy sets)becoming to be similar each other as progress of learning are unified. This unification causes the reduction of a number of parameters and learning time. Using fuzzy inference and making a rule with an appropriate state division, our proposed method makes it possible to construct a robust reinforcement learning system.

• Journal of Communications and Networks
• /
• 제11권5호
• /
• pp.509-517
• /
• 2009

This work proposes a new distributed and self-organized authentication scheme for mobile ad-hoc networks (MANETs). Apart from describing all its components, special emphasis is placed on proving that the proposal fulfils most requirements derived from the special characteristics of MANETs, including limited physical protection of broadcast medium, frequent route changes caused by mobility, and lack of structured hierarchy. Interesting conclusions are obtained from an analysis of simulation experiments in different scenarios.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2003년도 추계공동학술대회
• /
• pp.135-144
• /
• 2003

In this paper, the implementation techniques of intelligent nuclear material surveillance system based on the SOM(Self Organized Mapping) was described. Unattended continuous surveillance systems for nuclear facility result in large amounts of data, which require much time and effort to inspect. Therefore, it is necessary to develop system that automatically pinpoints and diagnoses the anomalies from data. In this regards, this paper presents a novel concept of a continuous surveillance system that integrates visual image and radiation data by the use of neural networks based on self-organized feature mapping

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2475-2480
• /
• 2003

A control system design based on coupled nonlinear oscillators (CNOs) for a self-organized swarm system is presented. In this scheme, agents self-organize to flock and arrange group formations through attractive and repulsive forces among themselves using CNOs. Virtual agents are also used to create richer group formation patterns. The objective of the swarm control in this paper is to follow a moving target with a final group formation in the shortest possible time despite some obstacles. The simulation results have shown that the proposed scheme can effectively construct a self-organized multi-agent swarm system capable of group formation and group immigration despite the emergence of obstacles.

• 한국재료학회지
• /
• 제11권12호
• /
• pp.1042-1046
• /
• 2001

Co-22%Cr alloy films are promising for high-density perpendicular magnetic recording media with their perpendicular anisotropy and large coercivity of 3000 Oe. We observed that a self organized nano structure(SONS) of fine ferromagnetic Co-enriched phase and paramagnetic Cr-enriched phase appears inside the grain of Co-Cr magnetic alloy thin films at the elevated substrate temperature after do-sputtering. We prepared 1000 $\AA$-thick Co-22%Cr films on 2000 $\AA$- SiO$_2$/Si(100) substrates at the deposition rate of 100 $\AA$/min with substrate temperatures of 3$0^{\circ}C$, 10$0^{\circ}C$, 15$0^{\circ}C$, 20$0^{\circ}C$, 30$0^{\circ}C$, and 40$0^{\circ}C$, respectively. We employed a vibrating sample magnetometer(VSM) to measure the B-H loops showing the saturation magnetifation, coercivity, remanence in in- plane and out- of- plane modes. In- plane coercivity, perpendicular coercivity, and perpendicular remanence increased as substrate temperature increased, how-ever they decreased after 30$0^{\circ}C$ slowly. Transmission electron microscope (TEM) characterization revealed that the self organized nano structure (SONS) appears at the elevated substrate temperature, which forms fine Co-enriched phases inside a grain, then it eventually affect the perpendicular magnetic property. Our results imply that we may tune the perpendicular magnetic properties with SONS obtained at appropriate substrate temperature.

• Biotechnology and Bioprocess Engineering:BBE
• /
• 제11권3호
• /
• pp.262-267
• /
• 2006

The objective of this study was to develop new self-organized nanogels as a means of drug delivery in patients with cancer. Pullulan (PUL) and deoxycholic acid (DOCA) were conjugated through an ester linkage between the hydroxyl group in PUL and the carboxyl group in DOCA. Three types of PUL/DOCA conjugates were obtained, differing in the number of DOCA substitutions (DS; 5, 8, or 11) per 100 PUL anhydroglucose units. The physicochemical properties of the resulting nanogels were characterized by dynamic light scattering, transmission electron microscopy, and fluorescence spectroscopy. The mean diameter of DS 11 was the smallest (approx. 100 nm), and the size distribution was unimodal. To determine the organizing behavior of these conjugates, we calculated their critical aggregation concentrations (CACs) in a 0.01-M phosphate buffered saline solution. They were $10.5{\times}10^{-4}mg/mL,\;7.2{\times}10^{-4} mg/mL,\;and\;5.6{\times}10^{-4} mg/mL$ for DS 5, 8, and 11, respectively. This indicates that DOCA can serve as a hydrophobic moiety to create self-organized nanogels. To monitor the drug-releasing behavior of these nanogels, we loaded doxorubicin (DOX) onto the conjugates. The DOX-loading efficiency increased with the degree of DOCA substitution. The release rates of DOX from PUL/DOCA nanogels varied inversely with the DS. We concluded that the PUL/DOCA nanogel has some potential for use as an anticancer drug carrier because of its low CAC and satisfactory drug-loading capacity.

• 한국콘텐츠학회논문지
• /
• 제13권11호
• /
• pp.28-36
• /
• 2013

본 논문은 손 형상 인식을 위한 보다 안정적이며 조명 변화와 회전에 강인하게 손 영역을 검출하며, 계산의 효율성과 검출 성능을 동시에 만족시키는 강인한 검출 알고리즘에 대해 제안한다. 제안한 알고리즘은 단일 카메라 환경에서 손 형상을 입력정보로 사용하여 전처리 과정을 거쳐 손 영역만을 분할한 후 자기조직화 특징 지도(SOFM: Self Organized Feature Map) 알고리즘을 이용하여 손 형상을 인식하게 된다. 그러나 조명 변화에 민감하고 자유도가 큰 손 영역을 정확히 인식하기란 쉽지 않으며 오차 범위도 크기 때문에 본 논문에서는 인식률을 높이기 위해 각각의 손 형상에 대한 회전 정보를 데이터베이스화 한 후 주성분 분석을 적용하여 군집화 함으로서 인식오차를 줄였다. 또한 차원 축소로 인해 많은 계산 량이 요구되지 않기 때문에 실시간 인식 시간도 줄일 수 있었다.