DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Park, Kyung-Soo | - |
dc.contributor.advisor | 박경수 | - |
dc.contributor.author | Jung, Gue-Woong | - |
dc.contributor.author | 정규웅 | - |
dc.date.accessioned | 2011-12-14T02:39:43Z | - |
dc.date.available | 2011-12-14T02:39:43Z | - |
dc.date.issued | 2003 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=181072&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/40546 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 산업공학과, 2003.2, [ vii, 97 p. ] | - |
dc.description.abstract | This thesis proposes an analysis method for the reliability of ring networks, especially Fiber Distributed Data Interface (FDDI) ring networks, and diagnosis methods for the crosstalk fault in optical switches, especially multistage interconnection networks. First, this thesis presents an analysis method for the K-terminal reliability of an FDDI ring network with both self-heal and station-bypass switches under the power-loss constraint. The analysis method uses an approach that combines a graph theoretic method and the consecutive k-out-olln:F system. The graph theoretic method is used to obtain the K-minimal Eulerian circuits, and the consecutive ksout-ofsn:F system is used to model the power-loss constraint. This approach provides K-terminal reliability in a closed-form expression, and the value of the reliability can be computed in a polynomial time. The computational results show that the effect of power-loss is not negligible, and this approach may be useful in evaluating the reliabilities of FDDI ring networks parametrically and making reliability comparisons. Second, this thesis suggests two methods to diagnose the crosstalk fault for optical multistage interconnection networks. One is a direct method, and the other is an indirect method. The direct method finds the crosstalk faults based on the crosstalk ratios that are computed using the power measurements of the crosstalks and signals. The indirect method finds the crosstalk faults by checking the crosstalks and the signals at the output of a network. Two methods are more efficient than the previous methods because they require much less tests than the previous methods. The key point of the two methods is the condition for satisfying the pure crosstalk constraint. The constraint means that the crosstalk generated from each switch is not combined in order to measure the pure crosstalk at the outputs of a network. The condition for satisfying the pure crosstalk constraint is expressed in a simple mathe... | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 링 네트워크 | - |
dc.subject | 망 신뢰도 | - |
dc.subject | 혼선 고장 | - |
dc.subject | 다단계 상호 연결망 | - |
dc.subject | 광 스위치 | - |
dc.subject | Network Reliability | - |
dc.subject | Ring Network | - |
dc.subject | Optical switch | - |
dc.subject | Crosstalk fault | - |
dc.subject | Multistage Interconnection Network | - |
dc.title | Network reliability analysis and fault diagnosis | - |
dc.title.alternative | 망 신뢰도 분석 및 고장 진단에 관한 연구 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 181072/325007 | - |
dc.description.department | 한국과학기술원 : 산업공학과, | - |
dc.identifier.uid | 000985341 | - |
dc.contributor.localauthor | Park, Kyung-Soo | - |
dc.contributor.localauthor | 박경수 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.