• 전자공학회논문지A
• /
• 제29A권7호
• /
• pp.10-18
• /
• 1992

In this paper, a simulation model for scope and filter parameters in CMIS/CMIP, OSI management protocol, has been discussed and the suitable scope type was represented by analyzing optimal polling time in network management system, queue size and response time. Scope and filter parameters in MGet, M-Set, M-Action, and =M-Delete services provide a scheme to select some managed objects, and the results are largely varied according to a scope type requested and the outcome of filtering. It has been simulated each scope type and all scope type mixed in a network by varying the offered load and a polling time. On the same time, the response time, queue size of the managing system and a delay time have been measured. As a results, the suitable scope type and polling time enhancing the network performance on the OSI network management is proposed.

• Journal of information and communication convergence engineering
• /
• 제17권3호
• /
• pp.185-190
• /
• 2019

The explosive growth of data traffic and services has created cost challenges for networks. Studies have attempted to effectively apply network slicing in fifth generation networks to provide high speed, low latency, and various compatible services. However, in network slicing using mixed-integer linear programming, the operation count increases exponentially with the number of physical servers and virtual network functions (VNFs) to be allocated. Therefore, we propose an efficient slice allocation method based on cluster technology, comprising the following three steps: i) clustering physical servers; ii) selecting an appropriate cluster to allocate a VNF; iii) selecting an appropriate physical server for VNF allocation. Solver runtimes of the existing and proposed methods are compared, under similar settings, with respect to intra-slice isolation. The results show that solver runtime decreases, by approximately 30% on average, with an increase in the number of physical servers within the cluster in the presence of intra-slice isolation.

• 산경연구논집
• /
• 제9권5호
• /
• pp.25-35
• /
• 2018

Purpose - The purpose of this paper is to partition a last-mile delivery network into zones and to determine locations of last mile delivery centers (LMDCs) in Bangkok, Thailand. Research design, data, and methodology - As online shopping has become popular, parcel companies need to improve their delivery services as fast as possible. A network partition has been applied to evaluate suitable service areas by using METIS algorithm to solve this scenario and a facility location problem is used to address LMDC in a partitioned area. Research design, data, and methodology - Clustering and mixed integer programming algorithms are applied to partition the network and to locate facilities in the network. Results - Network partition improves last mile delivery service. METIS algorithm divided the area into 25 partitions by minimizing the inter-network links. To serve short-haul deliveries, this paper located 96 LMDCs in compact partitioning to satisfy customer demands. Conclusions -The computational results from the case study showed that the proposed two-phase algorithm with network partitioning and facility location can efficiently design a last-mile delivery network. It improves parcel delivery services when sending parcels to customers and reduces the overall delivery time. It is expected that the proposed two-phase approach can help parcel delivery companies minimize investment while providing faster delivery services.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(1)
• /
• pp.113-116
• /
• 2002

It is required to network to make more intelligent packet processing and forwarding for increasing bandwidth and various services. Classification provides these intelligent to network which is acquired by increasing number of rules in classification rule set. In this Paper, we propose a classification algorithm efficient to scalable rule set ahead as well as Present small rule set. This algorithm has competition to existing methods by performance and advantage that it is mixed with another algorithm because il does not change original shape of rule set.

• 대한공간정보학회지
• /
• 제3권2호
• /
• pp.91-100
• /
• 1995

본 논문은 위성탐사 이미지의 분류에서 한 화소(SPOT HRV의 밴드 1-3의 경우 $200{\times}20m$)에 포함된 혼합된 정보의 분류를 시도한다. 먼저 기존의 분류기법에서 발생되는 정보의 손실과 혼합화소에 내포된 정보의 불확실성에 대해 알아보고 이를 해결하기 위한 방법으로 피지 시그모이드 함수와 역전파 신경망을 이용한 기법을 제안하며, 이를 실험하고 비교 분석한다.

• 한국정보시스템학회지:정보시스템연구
• /
• 제17권3호
• /
• pp.39-58
• /
• 2008

Mixed Chinese Postman Problem (MCPP) is a practical generalization of the classical Chinese Postman Problem (CPP) and it could be applied in many real world. Although MCPP is useful in terms of reality, MCPP has been proved to be a NP-complete problem. To find optimal solutions efficiently in MCPP, we can reduce searching space to be small effective searching space containing optimal solutions. We propose GENIIS methodology, which is a kind of hybrid algorithm combines the approximate algorithms and genetic algorithm. To get good solutions in the effective searching space, GENIIS uses approximate algorithm and genetic algorithm. This paper validates the usefulness of the proposed approach in a simulation. The results of our paper could be utilized to increase the efficiencies of network and transportation in business.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.21-23
• /
• 2022

ARM 아키텍처를 사용하는 임베디드 시스템에서 int8, fp16, fp32 데이터를 조합하여 c/c++로 작성된 mixed precision CNN 을 실행시키기 위한 프레임워크 구성으로, 네트워크의 레이어마다 다른 정밀도를 사용하여 네트워크 경량화 및 추론 정확도 향상을 위한 최적의 설정을 탐색하는 실험 및 분석이 가능토록 하는 것을 목적으로 한다. 주요 구성은 network forwarding 중 레이어의 입력이 레이어에 설정된 정밀도와 다를 경우 실행되는 양자화/반양자화를 c/c++로 바인딩된 tensorflow 의 quantization 모듈을 사용하여 진행하고 ARM 시스템에서 c/c++의 fp16 을 사용하기 위해 fp16 를 컴파일이 가능한 ARM compiler 를 사용하는 프레임워크를 제안한다.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.858-869
• /
• 2023

As one of the pivotal techniques of image restoration, single-channel blind source separation (SCBSS) is capable of converting a visual-only image into multi-source images. However, image degradation often results from multiple mixing methods. Therefore, this paper introduces an innovative SCBSS algorithm to effectively separate source images from a composite image in various mixed modes. The cornerstone of this approach is a novel triple generative adversarial network (TriGAN), designed based on dual learning principles. The TriGAN redefines the discriminator's function to optimize the separation process. Extensive experiments have demonstrated the algorithm's capability to distinctly separate source images from a composite image in diverse mixed modes and to facilitate effective image restoration. The effectiveness of the proposed method is quantitatively supported by achieving an average peak signal-to-noise ratio exceeding 30 dB, and the average structural similarity index surpassing 0.95 across multiple datasets.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2002년도 추계학술대회 논문집(I)
• /
• pp.484-489
• /
• 2002

Whenever the accidents occur in the railway areas, their prompt recovery is very important. The individual way has been used to resolve these accidents by direct visiting the site and reporting to others who cares for these cases. As a result, we are planning to operate the MTS (Moving picture Transmission System) for a timely information's transmission to the related peoples in the central control center by transforming the picture information on the accident-site to the digital information. The proposed system is not a transmission system using exclusive lines, but utilizing the existing railway computer network. We construct a network server device so that a dispatcher can easily connect with the server through the railway intranet. In some railway environments such as disturbed field situations and geographical condition(in tunnels and bridges, etc) building up the wired network is difficult. Solving this problem, we reviewed the wireless network. Finally we proposed the mixed wireless network that is able to cover the wired network. We then installed and tested the wired network and wireless network respectively in wayside of a railway field. Through the testing, we identified many detailed matters for some network construction approaches to the control room and many field device-connecting schemes. This research will contribute to minimize the number of staff dispatched to a disturbed field, decrease the accident recovery time and improve the quality of service to the passenger.

• 한국정보시스템학회지:정보시스템연구
• /
• 제29권1호
• /
• pp.267-287
• /
• 2020

Purpose Online platform companies can increase the spread of content by communicating with users who have diverse preferences through social networks. Previous studies show the mixed effect on the network overlap, and there was a limited examinations for the underlying mechanism. This study expects high academic and practical implications that can be provided by studying on the user's viewership network. The purpose of this research is to examine the effects of network overlap on the users' viewership for creators of user-generated content in YouTube. We explain the direct and in-direct effects through the content sharing and the valence of user ratings. Design/methodology/approach The data contains 45 channels and 4,085 video clips from YouTube. We control the effect of the categories, channel characteristics, and vide clip characteristics on the viewership. PROCESS macro were used to analyze the direct and in-direct effects of network overlap. Findings The analysis results showed that the network overlap directly affect on the users' viewership. The variable decreases the moderators (i.e., content sharing and the valence of user ratings). This result implies that the users can not satisfy their need for uniqueness which is achieved by content sharing and rating in the overlapped network.