• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.39 no.11
• /
• pp.1-11
• /
• 2002

To increase network survivability and to guarantee reliable services in MPLS networks, when failures occur either at any link or node of a working path, traffics on the failed path should be switched to other restoration paths. In this paper, we propose a path restoration method using duplication of working paths for link failure and its associated failure detection method, which utilizes OAM cells when all the duplicated links or nodes of the path failed. We analyze the performance of the proposed restoration method and show that especially the method is very efficient to restore high priority traffics. Also, through simulation we prove that the proposed restoration method is superior in performance to several existing methods suggested so far.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.1
• /
• pp.245-251
• /
• 2000

파이프라인 프로세서를 위한 이중 경로 전략의 성능을 개선하기 위해, 본 논문에서는 통합 신뢰 매커지즘과 적극적 이중경로 전략(EDPS)을 제안한다. 통합 신뢰 매커니즘은 동적 신뢰 매커니즘과 정적 신뢰 매커니즘을 결합한 것으로 기존의 신뢰 매커니즘보다 신뢰 예측 정확도를 높일 수 있고 제안하는 EDPS와 결합하여 사용한다. EDPS는 높은 신뢰 집합에 g속하는 분기 명령어도 가능한 경우에는 두 경로를 모두 사용하여 조건 분기 명령어로 인해 발생하는 분기 지연의 총합을 줄일 수 있다. 6개 벤치마크에 대한 추적 기반의 시뮬레이션을 통해, 제안된 통합 신뢰 매커니즘을 사용하는 EDPS가 기존의 선택적 이중 경로 실행에 비해 분기 지연의 총합을 22%을 줄일 수 있다.

• Proceedings of the Optical Society of Korea Conference
• /
• 2003.02a
• /
• pp.142-143
• /
• 2003

EDFA의 이득(gain)과 잡음지수(noise figure)는 매우 중요한 특성 중의 하나이다. 전송 시스템에서 필연적으로 발생하는 ASE 잡음을 제거하는 것이 중요하다. EDFA의 중간에 필터(filter)와 아이솔레이터(isolator)를 삽입하여 불필요한 ASE를 제거함으로써 이득과 잡음지수를 향상시킬 수 있으며, 기존의 단일경로(single-pass) EDFA에서 필터의 위치가 40-60％ 근처에서 이득과 잡음지수 향상이 가장 크게 된다. 그러나, 거울(mirror)을 사용하여 증폭 효율을 높인 이중경로(double-pass) EDFA에서 필터에 의한 이득 향상에 관한 연구발표는 아직까지 없다. (중략)

• The KIPS Transactions:PartA
• /
• v.12A no.6 s.96
• /
• pp.557-562
• /
• 2005

Throughput and latency of interconnection network are important factors of the performance of multiprocessor systems. The dual-link CC-NUMA architecture using point-to-point unidirectional link is one of the popular structures in high-end commercial systems. In terms of optimal path between nodes, several paths exist with the optimal hop count by its native multi-path structure. Furthermore, transaction latency between nodes is affected by congestion of links on the transaction path. Hence the transaction latency may get worse if the transactions make a hot spot on some links. In this paper, I propose a dynamic transaction routing algorithm that maintains the balanced link utilization with the optimal path length, and I compare the performance with the fixed path method on the dual-link CC-NUMA systems. By the proposed method, the link competition is alleviated by the real-time path selection, and consequently, dynamic transaction algorithm shows a better performance. The program-driven simulation results show $1{\~}10\%$ improved fluctuation of link utilization, $1{\~}3\%$ enhanced acquirement of link, and $1{\~}6\%$ improved system performance.

• Proceedings of the Optical Society of Korea Conference
• /
• 2003.02a
• /
• pp.140-141
• /
• 2003

광섬유 통신에서 넓은 대역폭 증폭을 이루기 위하여 광대역 증폭기가 필요하다. 증폭기의 대역폭을 넓히기 위해 C-대역과 L-대역 EDFA를 결합하여 사용하는 등의 여러 가지 방법이 제안되고 있다. 그 중 L-대역을 결합하는 방법이 실제 시스템 적용에서 가장 즉시 이용가능한 방법이라 할 수 있다. L-대역 EDFA의 경우 상대적으로 이득이 낮으며 효율이 떨어진다. 이득이나 효율을 높이기 위해서 FBG, 양방향 펌핑(bidirectional pump), 그리고 이중경로(double-pass) 구조를 이용하는 방법이 제안되었다. (중략)

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.11
• /
• pp.5885-5889
• /
• 2013

In general, the network survivability means that The user do not know the network faults and the recovery of those. For this, we use the dual(multi) routes between each two nodes. It is important that the each dual routes have efficient costs(or minimum). Even if one route is the minimum cost in case of no fault, another route of dual may be very large cost in case of fault case. Therefore we need the dual routes of each two nodes having the efficient(or minimum) costs. In this paper we find the network design method for the dual routes of each two node having the efficient costs. Although the design method is very simple and heuristic and it may be not useful for some networks, we will use it in various network environment.. Because this design method can be used very easy. A sample design will proof this usefulness.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2012.10a
• /
• pp.252-254
• /
• 2012

물류센터내에서는 작업자에 의한 지게차 작업이 주를 이루고 있으나, 점차 물류센터의 대형화, 자동화 및 지능화에 의해 무인지게차에 의한 작업에 관한 연구가 진행되고 있다. 최근 작업 생산성을 높이기 위해, 본 연구진에 의해 이중 포크 (Dual Fork) 장착형 무인지게차에 대한 검토가 이루어지고 있으며, 이중 포크형 무인지게차에 대해서는 새로운 방법의 동적경로할당이 요구된다. 본 연구에서는 4단계 심플한 동적 경로 할당 알고리즘을 제안하였으며, 실제 물류센터에서 유용하게 사용될 수 있도록 하였다. 마지막으로는 시뮬레이션을 통하여 그 유효성을 검증하였다.

• Journal of Internet Computing and Services
• /
• v.4 no.2
• /
• pp.1-9
• /
• 2003

Path restoration, which is needed to increase network survivability in MPLS networks, is a technology of supplying reliable service by delivering the traffics through restoring the paths when links or nodes of a working path are failed. In this paper, we propose a path restoration method using duplication of working paths for link failure as a method of restoring paths for the high priority traffics. The existing path restoration method transmits the traffics to the path restoration uniformly in regardless of the priority while link obstacle. It has a problem to take a long time to restore the connection of the high priority traffics. On the other hand, the suggested restoration method establishes the duplication path for the working path, which transmits the high priority traffics and transmit the traffics through the duplication path. It has a strong point to restore the connection quickly. Also, through simulation we analyze the performance of the restoration system using the duplication path and prove that the proposed restoration method is superior in performance to the existing restoration methods regarding to the high priority traffics.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.1
• /
• pp.111-118
• /
• 2015

This paper deals with the routing and wavelength assignment problem (RWAP) that decides the best lightpaths for multiple packet demands for (s,t) in optical communication and assigns the minimum number of wavelengths to given lightpaths. There has been unknown of polynomial-time algorithm to obtain the optimal solution for RWAP. Hence, the RWAP is classified as NP-complete problem and one can obtain the approximate solution in polynomial-time. This paper decides the shortest main and alternate lightpath with same hop count for all (s,t) for given network in advance. When the actual demands of communication for particular multiple packet for (s,t), we decrease the maximum utilized edge into b utilized number using these dual-paths. Then, we put these (s,t) into b-wavelength bins without duplicated edge. This algorithm can be get the optimal solution within O(kn) computational complexity. For two experimental data, the proposed algorithm shows that can be obtain the known optimal solution.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.4
• /
• pp.833-839
• /
• 2007

In case of wireless sensor networks being applied to a target environment having continuous data transfer, high-speed data transmission and processing is necessary to meet increased bandwidth needs. This paper proposes a novel multipath routing algorithm to increase the bandwidth of continuous data stream for the wireless sensor network using dual channel routers. We evaluate performance of the proposed algorithm in the wireless sensor network using the dual channel touter based on ZigBee protocol. According to the experimental evaluation, the proposed algorithm showed outstanding performance enhancement with the bandwidth of 150kbps, which is about twice of existing single channel router.