• 산업경영시스템학회지
• /
• 제4권4호
• /
• pp.63-69
• /
• 1981

As enterprises are getting bigger and bigger and more competecious, an engineering economy for the maximization of profit based on basic theory must be considered. This thesis present dynamic computer model for the decision which controls complicated and various man- machine system optimally. This model occur in general stage can be adaptable to every kind of enterprises. So, any one who has no expert knowledge is able to get the optimal solution. And decision tree used in this paper can be applied in every kinds of academic circles as well as whole the industrial world. This paper studied optimal management of engineering project based upon basic theory of engineering economy. It introduces and functionizes the variables which generalize every possible elements, set up a model in order to find out the variable which maximize the calculated value among many other variables. And the selected values ate used as decision- marking variables for the optimal management of engineering projects. It found out some problem of this model. They are : 1. In some kinds of man-machine system it refers to Probability, but other case, it depends on only experimental probability. 2. Unless decision making process (decision tree) goes on, this model can not be applied. So these cases, this paper says, can be solved by adapting finite decision tree which is analyzed by using the same technic as those in product introduction problem. And this paper set up the computer model in order to control every procedure quickly and optimally, using Fortran IV.

• 한국지능시스템학회논문지
• /
• 제15권5호
• /
• pp.626-631
• /
• 2005

본 논문에서는 Dynamic Programming(DP)을 적용한 트리구조 미로내의 목표물 탐색 알고리즘을 구현한다. DP는 큰 문제를 이루는 작은 문제들을 먼저 해결하고 작은 문제들의 최적해를 이용하여 순환적으로 큰 문제를 해결한다. 먼저 실험을 위해 적외선 센서론 부착한 소형 이동 로봇과, 'Y'형태로 갈라진 길을 연결한 트리 구조의 미로 환경을 구성한다. 실험에서는 두 개의 서로 다른 알고리즘 - 좌수법, DP - 을 사용하여 목표물 탐색을 시도한다. 마지막으로 위 실험을 통해 DP를 미로 탐색문제에 적용했을 때의 성능을 검증한다.

• 한국컴퓨터정보학회논문지
• /
• 제11권3호
• /
• pp.73-78
• /
• 2006

기존의 문자 영역 추출 방법은 전체 영상으로부터 컬러 영역 분할이나 프레임 차 방법을 이용하였다. 이들 방법은 휴리스틱에 많이 의존하므로 추출하려는 문자의 사전 정보를 가지고 있어야한다는 점과 구현에 많은 어려움이 존재한다. 본 논문에서는 휴리스틱한 부분을 줄이고 알고리즘을 단순화한 방법을 제안하고자 한다 문자의 지형학적 특징점을 추출하고 이 점들을 MST(Minimum Spanning Tree)를 형성하여 문자의 후보 영역을 추출한다. 문자 영역을 후보 영역의 검증을 통하여 추출한다. 실험 결과 문자의 후보 영역 추출율은 100%이었으며 최종 문자 영역 추출율은 98.2%이었다. 또한 복잡한 영상에서 존재하는 문자 영역도 잘 추출됨을 볼 수 있다.

• 제어로봇시스템학회논문지
• /
• 제4권1호
• /
• pp.123-133
• /
• 1998

In this paper, an algorithm of fault detection and diagnosis during operation of induction motors under the condition of various loads and rates is investigated. For this purpose, the spectrum pattern of input currents is used in monitoring the state of induction motors, and by clustering the spectrum pattern of input currents, the newly occurrence of spectrum patterns caused by faults are detected. For the diagnosis of the fault detected, a fuzzy fault tree is designed, and the fuzzy relation equation representing the relation between an induction motor fault and each fault type, is solved. The solution of the fuzzy relation equation shows the possibility of occurence of each fault. The results obtained are summarized as follows : (1) Using clustering algorithm by unsupervised learning, an on-line fault detection method unaffected by the characteristics of loads and rates is implemented, and the degree of dependency for experts during fault detection is reduced. (2) With the fuzzy fault tree, the fault diagnosis process become systematic and expandable to the whole system, and the diagnosis for sub-systems can be made as an object-oriented module.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권5호
• /
• pp.1303-1315
• /
• 2012

In the draft of the IEEE 802.11s standard, a tree topology is established by the proactive tree-building mode of the Hybrid Wireless Mesh Protocol (HWMP). It is used for cases in which the root station (e.g., gateway) is an end point of the majority of the data connections. In the tree topology, the root or central stations (e.g., parent stations) are connected to the other stations (e.g., leaves) that are one level lower than the central station. Such mesh stations are likely to suffer heavily from contention in bottleneck links when the network has a high traffic load. Moreover, the dependence of the network on such stations is a point of vulnerability. A failure of the central station (e.g., a crash or simply going into sleep mode to save energy) can cripple the whole network in the tree topology. This causes performance degradation for end-to-end transmissions. In a connected mesh topology where the stations having two or more radio links between them are connected in such a way that if a failure subsists in any of the links, the other link could provide the redundancy to the network. We propose a scheme to utilize this characteristic by organizing the network into concentric tiers around the root mesh station. The tier structure facilitates path recovery and congestion control. The resulting mode is referred to as Tier-based Proactive Path Selection Mode (TPPSM). The performance of TPPSM is compared with the proactive tree mode of HWMP. Simulation results show that TPPSM has better performance.

• 한국정보과학회논문지:데이타베이스
• /
• 제34권3호
• /
• pp.217-231
• /
• 2007

이동 컴퓨팅(Mobile Computing)환경에서 빈번한 접속 단절은 클라이언트의 캐시 일관성(Consistency)문제로 직결된다. 이러한 캐시 일관성 문제를 해결하기 위해 무효화 보고(Invalidation Report)에 관한 연구가 진행되고 있다. 그러나 기존의 무효화 보고 기법은 서버 측의 데이타의 수가 많아지거나, 갱신(Update)되는 데이타의 양이 증가하게 되면 무효화 보고 기법의 크기의 증가 및 캐시 효율성의 저하를 가지고 온다. 또한 캐시 전체의 무효화를 확인하는 보고 기법은 선택적 청취(Selective tuning)를 지원하지 못한다. 본 논문에서는 이러한 기존 방식의 문제점을 해결하며 효율성을 개선한 무효화 보고 기법으로써 TTCI(Timestamp Tree-based Cache Invalidation scheme)를 제안한다. 갱신된 데이타의 타임스탬프를 이용하여 타임스탬프 트리를 구성하고 데이타의 아이디를 갱신된 순서대로 나열하여 무효화 보고를 만든다. 이를 활용하게 되면 각 클라이언트는 자신의 단절(Disconnect)된 시점에 맞는 정보를 확인하여 캐시의 이용률을 증가 시킬 수 있다. 또한 트리 구조를 이용하여 선택적 청취를 가능하게 함으로써 클라이언트의 자원 소비를 줄일 수 있다. 이러한 본 구조의 효율성을 검증하기 위해 기존의 기법인 DRCI(Dual-Report Cache Invalidation)와 성능비교를 실시한다.

• 한국철도학회논문집
• /
• 제9권3호
• /
• pp.264-270
• /
• 2006

In this paper we present a safety plan to be applied to the development of the TCS(Train Control System). The safety plan that can be applied to the life cycle of a system, from the conceptual design to the dismantlement, shows the whole process of the paper work in detail through the establishment of a goal, analysis and assessment, the verification. In this paper we study about the making a plan, the preliminary hazard analysis, the hazard identification and analysis to guarantee the safety of the TCS. The process far the verification of the system safety is divided into several steps based on the target system and the approaching method. The guarantee of the system safety and the improvement of the system reliability is fellowed by the recommendation of the international standards.

• 한국제4기학회지
• /
• 제6권1호
• /
• pp.21-32
• /
• 1992

The paleoclimate of the Chungbu Mountainous Region, Mts. Seorak and Sobaek, was estimated by means of dendroclimatic methodology, The annual growth value of tree-rings is deeply interrelated with the mean temperature of April-May and July-August. The mean temperature of April-May of the reconstruction period(1635-1911), observation period(1912-1989), and the whole period(1635-1989) is 13.58$^{\circ}C$, 13.69$^{\circ}C$, and 13.6$0^{\circ}C$, respectively. That of July-August is $24.50^{\circ}C$, $24.62^{\circ}C$, and $24.58^{\circ}C$ respectively. The reconstructed mean temperature data for April-May and for July-August reveal 13.2 and 12.9 year cycles.

• The Korean Journal of Ecology
• /
• 제19권3호
• /
• pp.223-230
• /
• 1996

Five measurements of species diversity (richness index, diversity index, evenness index, dominance index and species sequences-importance curve) and their relationships with sltitude, tree age and community type were studied in $T\v{o}gyusan$ National Park, Korea. Altitude and tree age were the major variables explaining the differences of species diversity in the whole forest. Species richness index, diversity index and evenness index of Quercus mongolica and Pinus densiflora communities were lower than those of Carpinus laxiflora, Q. serrata, Fraxinus mandshurica, Q. mongolica - P. densiflora communities than that in the other communities. The species sequence-importance curve of the forest communities in $T\v{o}gyusan$ coincided with the ideal curve calculated by the lognormal-distribution theory.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.906-909
• /
• 2000

Conventional caption extraction methods use the difference between frames or color segmentation methods from the whole image. Because these methods depend heavily on heuristics, we should have a priori knowledge of the captions to be extracted. Also they are difficult to implement. In this paper, we propose a method that uses little heuristics and simplified algorithm. We use topographical features of characters to extract the character points and use KMST(Kruskal minimum spanning tree) to extract the candidate regions for captions. Character regions are determined by testing several conditions and verifying those candidate regions. Experimental results show that the candidate region extraction rate is 100%, and the character region extraction rate is 98.2%. And then we can see the results that caption area in complex images is well extracted.