• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.340-340
• /
• 2019

지구온난화로 유발되는 기후변동성이 증가함에 따라서 정확한 수문인자의 예측은 전 세계적으로 주요 관심사항이 되고 있다. 최근에는 고성능 컴퓨터 자원의 증가로 수문기상학 연구에서 동일한 학습량에 비하여 정확도의 향상이 뚜렷한 기계학습 구조를 활용하여 위성영상 기반의 대기예측, 태풍위치 추적 및 강수량 예측 등의 연구가 활발하게 진행되고 있다. 본 연구에는 기계학습 중 시계열 분석에 널리 활용되고 있는 순환신경망(Recurrent Neural Network, RNN) 기법의 대표적인 LSTM(Long Short-Term Memory) 네트워크를 이용하여 수문인자를 예측하였다. LSTM 네트워크는 가중치 및 메모리 요소에 대한 추가정보를 셀 상태에 저장하고 시계열의 길이 조정하여 모형의 탄력적 활용이 가능하다. LSTM 네트워크를 이용한 다양한 수문인자 예측결과 RMSE의 개선을 확인하였다. 따라서 본 연구를 통하여 개발된 기계학습을 통한 수문인자 예측기술은 권역별 수계별 홍수 및 가뭄대응 계획을 능동적으로 수립하는데 활용될 것으로 판단된다. 향후 연구에서는 LSTM의 입력영역을 Bayesian 추론기법을 활용하여 구성함으로 학습과정의 불확실성을 정량적으로 제어하고자 한다.

• Journal of Korea Port Economic Association
• /
• v.38 no.4
• /
• pp.25-43
• /
• 2022

Unlike the past, efforts must be made to interpret physical distribution from a network perspective as the service area expands spatially. In addition, logistics networks are undergoing rapid changes due to various changes in the environment. Therefore, the purpose of this study is to analyze the changes in the structure of maritime cargo and the centrality of ports using social network analysis. Using the trade data of domestic maritime at five-year intervals, we investigated changes in the network structure and identified the main factors that affect the centrality of domestic ports. Ports with the highest centrality, which is seen as a port that plays the role of an intermediary, emerged in the order of Busan and Ulsan. This study predicts patterns of domestic cargo trade over the next 20 years based on changes in port centrality and understanding of maritime cargo network, and can be used as reference materials for risk preparation.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.11 s.353
• /
• pp.78-84
• /
• 2006

This paper proposes a method to endow the mobility to the Zigbee protocol which is a short distance wireless communication technology among the standards of wireless sensor networks. The Zigbee has a low price, long life battery, simple structure and connectivity. So it is a suitable standard for the short distance wireless communication of $10m{\sim}70m$. As the Zigbee does not support the mobility, additional factors should be studied to deal adeptly with the diverse change in the real field. By solving the problem that the previous connection and binding table of the Zigbee end device should be reset due to the Zigbee end device address variation when the Zigbee end device moves from the present router to another router, this paper presents the method to communicate without data loss or disconnection.

• The Journal of the Korea Contents Association
• /
• v.11 no.11
• /
• pp.137-144
• /
• 2011

This study explored the risk sharing mechanism in the US TV drama industry. The hollywood system of TV production and distribution can be understood to be the result of the efforts to reduce the uncertainty and the risk of investment. Since the 1990s, the vertical integration of major studios and networks has been one of the core strategies for risk sharing. In the strategy, the most important role of networks is to schedule new series that the affiliated studios produce. Networks also provide the new series their brand value. On the other hand, the most important role of studios in the vertical integrated environment is to provide the affiliated networks the infrastructures for production and distribution, financial management, and product management. In addition, networks keep the right on the series in the secondary markets, and attempt to increase the aggressive investment on new series on the base of the rights. However, such a risky investment can be the potential burden and risk for the networks in the future. This paper shows the importance of the reasonable risk sharing in the television drama industry to the Korean TV drama industry that has experienced the conflict between networks and independent production companies.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.6
• /
• pp.137-144
• /
• 2018

Vehicular Ad-hoc Networks (VANETS) have enabled to provide a variety of applications such as accident notification, content usage, etc. These applications have spatio-temporal data which have an interesting region and a lifetime according to their properties. However, geocasting protocols to deliver data to an interesting region can provide data to all vehicles in the region through a single transmission only at the current time, but cannot provide data to vehicles passing through the region during the lifetime of the data. Thus, we propose a spatio-temporal geocasting protocol called STGP using a regional caching scheme to send data to vehicles in an interesting region during a data lifetime in VANETs. For efficient and reliable regional caching, the proposed protocol uses the beacon-based data sharing, the extra caching elimination, and the distance-based caching exchange. Simulation results verify that the proposed protocol achieves more reliable and efficient data delivery compared with the existing protocol.

• The KIPS Transactions:PartB
• /
• v.15B no.2
• /
• pp.137-146
• /
• 2008

Fault localization plays a significant role in enormous distributed system because it can identify root cause of observed faults automatically, supporting self-managing which remains an open topic in managing and controlling complex distributed systems to improve system reliability. Although many Artificial Intelligent techniques have been introduced in support of fault localization in recent research especially in increasing complex ubiquitous environment, the provided functions such as diagnosis and prediction are limited. In this paper, we propose fault localization for self-managing in performance evaluation in order to improve system reliability via learning and analyzing real-time streams of system performance events. We use probabilistic reasoning functions based on the basic Bayes' rule to provide effective mechanism for managing and evaluating system performance parameters automatically, and hence the system reliability is improved. Moreover, due to large number of considered factors in diverse and complex fault reasoning domains, we develop an efficient method which extracts relevant parameters having high relationships with observing problems and ranks them orderly. The selected node ordering lists will be used in network modeling, and hence improving learning efficiency. Using the approach enables us to diagnose the most probable causal factor with responsibility for the underlying performance problems and predict system situation to avoid potential abnormities via posting treatments or pretreatments respectively. The experimental application of system performance analysis by using the proposed approach and various estimations on efficiency and accuracy show that the availability of the proposed approach in performance evaluation domain is optimistic.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.2
• /
• pp.48-59
• /
• 2024

As functions that support virtualization on their own in hardware are developed, user applications having various workloads are operating efficiently in the virtualization system. SR-IOV is a virtualization support function that takes direct access to PCI devices, thus giving a high I/O performance by minimizing the need for hypervisor or operating system interventions. With SR-IOV, network I/O acceleration can be realized in virtualization systems that have relatively long I/O paths compared to bare-metal systems and frequent context switches between the user area and kernel area. To take performance advantages of SR-IOV, network resource management policies that can derive optimal network performance when SR-IOV is applied to an instance such as a virtual machine(VM) or container are being actively studied.This paper evaluates and analyzes the network performance of SR-IOV implementing I/O acceleration is compared with Virtio in terms of 1) network delay, 2) network throughput, 3) network fairness, 4) performance interference, and 5) multi-network. The contributions of this paper are as follows. First, the network I/O process of Virtio and SR-IOV was clearly explained in the virtualization system, and second, the evaluation results of the network performance of Virtio and SR-IOV were analyzed based on various performance metrics. Third, the system overhead and the possibility of optimization for the SR-IOV network in a virtualization system with high VM density were experimentally confirmed. The experimental results and analysis of the paper are expected to be referenced in the network resource management policy for virtualization systems that operate network-intensive services such as smart factories, connected cars, deep learning inference models, and crowdsourcing.

• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.3
• /
• pp.138-146
• /
• 2010

It is well known that many biological networks are very robust against various types of perturbations, but we still do not know the mechanism of robustness. In this paper, we find that there exist a number of feedback loops in a real biological network compared to randomly generated networks. Moreover, we investigate how the topological property affects network robustness. To this end, we properly define the notion of robustness based on a Boolean network model. Through extensive simulations, we show that the Boolean networks create a nearly constant number of fixed-point attractors, while they create a smaller number of limit-cycle attractors as they contain a larger number of feedback loops. In addition, we elucidate that a considerably large basin of a fixed-point attractor is generated in the networks with a large number of feedback loops. All these results imply that the existence of a large number of feedback loops in biological networks can be a critical factor for their robust behaviors.

• Electronics and Telecommunications Trends
• /
• v.30 no.4
• /
• pp.92-101
• /
• 2015

미래 전장은 정보 지식 기반의 첨단 전력체계를 확충하기 위해 향후 전력구조를 통합, 지휘통제통신(C4I) 체계와 생존성과 통합성이 향상된 전장의 네트워크중심전(NCW) 수행능력을 향상시킬 것이다. 사이버물리시스템(Cyber-Physical Systems: CPS)은 함정전투체계에 적용되고 있는 DDS를 포함하여 국방 M&S의 근간인 Live, Virture, Constructive(L-V-C) 체계의 큰 축을 형성하고 있다. 사물인터넷(Internet of Things: IoT) 기술은 센서네트워크, 통신, Radio Frequency Identification(RFID), Ubiquitous Sensor Network(USN), Machine to Machine(M2M), D2D 기술 및 상황인지, 지능서비스를 위한 정보수집/가공/융합/분석/예측기술을 포괄적으로 포함한 기술로서 미래산업을 이끌어 갈 차세대 선도 기술이며, 특히 군사적으로도 감시정찰 센서네트워크(USN), 견마형로봇, 경전투로봇과 무인기 기술 및 전술정보통신망체계(TICN) 등 첨단 통신네트워크 기술의 전력화 추세는 IoT 기술의 적용영역을 넓혀주고 있다. 감시정찰체계(Sensor)에서는 감시정찰 분야 영상정보 처리, 표적탐지 등과 관련된 IoT 기술 소요와 지휘통제통신(C4I) 체계의 상호운용성, 데이터링크, 지능형 통신체계 등 C4I 관련 IoT 기술 소요 및 타격체계(Shooter)의 내장형 SW 등 유 무인 무기체계 관련 IoT 기술의 소요가 증대될 것으로 예상된다. 본고는 CPS 및 IoT 기술의 군사적 활용방안 및 획득전략에 대한 적용기술 및 발전방향을 살펴본다.

• Electronics and Telecommunications Trends
• /
• v.29 no.6
• /
• pp.71-81
• /
• 2014

5G 무선통신시스템은 동일한 영역에서 스펙트럼 사용 효율성을 개선하기 위해 매크로셀과 소형셀이 공존하는 이종 네트워크(HetNet: Heterogeneous Network) 형태로 진화하고 있으며, 급증하는 모바일 트래픽을 효율적으로 처리하기 위해 소형셀들을 고밀도 네트워크(High dense network)로 구축하는 방안이 연구되고 있다. 매크로셀과 고밀도 소형셀들이 중첩되어 구축되는 HetNet 기반 셀룰러 네트워크에서 소형셀 시스템의 구성과 파라미터 최적화를 통한 성능 유지를 운영자가 수동으로 조정하는 것은 한계가 있으므로 네트워크 환경변화에 따라 시스템에서 자율적으로 파라미터를 조정하여 시스템 성능을 유지하는 기술이 요구되고 있다. 본고에서는 시스템 운용 중 자율적인 최적화를 통해 시스템 성능을 최적으로 유지하고 유지비용을 최소화하는 3GPP 자율적 네트워크 최적화 기술을 소개한다.