• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.11-14
• /
• 2022

본 논문에서는 회귀 문제에서 예측값들의 분산을 줄이기 위한 딥뉴럴 네트워크 구조를 제안한다. 일반적인 회귀 문제에서 딥뉴럴 네트워크 학습 시, 하나의 입력에 대한 레이블 값을 이용하여 학습한다. 본 눈문에서는 하나의 입력에 대한 레이블 값뿐만 아니라 두 입력에 대한 레이블 값들의 차이를 학습시키는 딥뉴럴 네트워크 구조를 제안한다. 통계학 이론을 통하여 예측값들의 분산이 줄어든다는 것을 증명한다. 또한, 배관 곡관의 감육두께를 예측하는 문제를 통해 제안된 네트워크의 성능을 검증한다. 일반적인 딥뉴럴 네트워크 구조를 이용하였을 때에 비하여 제안한 네트워크 구조를 이용하였을 때, 회귀 문제의 예측값들의 분산이 감소함을 확인한다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.4
• /
• pp.679-686
• /
• 2020

With global data totals expected to grow to 175 ZB by 2025, data processing capabilities in the network area are becoming more important. In particular, data centers are becoming more stubborn as data usage increases, and the load generated by incoming data should be appropriately distributed to reduce costs and efficiently operate. The SDN technology, which emerged to overcome the limitations of the existing network system, removes rigidity of the Legacy system by separating H/W and S/W from the network equipment, and can be effectively applied to load balancing in the data center using S/W-based flexibility. In this paper, we propose ways to increase efficiency by classifying data received from users by type by utilizing SDN technology, and transmitting and processing classified data in order of response speed of servers in the data center.

• The Journal of Korean Association of Computer Education
• /
• v.15 no.5
• /
• pp.55-62
• /
• 2012

With the development of mobile communications and Internet technology, there is a strong need to provide seamless and fast connectivity for roaming devices. Generally, the mobile host (MH) may have several available networks when entering a new wireless area. However, the standard of decision for user's internet connection is provided only the subsystem identification (SSID) and signal strength of access point (AP). These two standards could not enough to decide optimal AP to the MH. Therefore, to decide the optimal AP, more information is needed. In this paper, we present additional information such as status of MH, capacity, current load, and depth of network hierarchy, by router advertisement message at layer 3. Also, we proposed decision engine (DE) on the MH that analyzes APs and decides the optimal AP automatically by AP's status information. For the MH, wireless connection period is increased, the power consumption is decreased, and the signaling overhead is reduced. For AP and router, the load balancing is provided and the network topology can also be more efficient.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.3B
• /
• pp.168-174
• /
• 2007

All of users who connect to mobile game want to play seamless real-time game without any loads on game servers. However, as for existing methods, it is not available to effectively distribute server loads. Therefore, in this paper, we design and implement an uniform variance-based load balancer for distributing loads of game servers for wireless online games by suitable load balancing methods. With this methods, we can deal with dynamically increasing game servers as the information about load balancing of specific period is applied to the administrator.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04a
• /
• pp.643-645
• /
• 2002

이동 에이전트는 분산 응용에 있어서 네트워크의 부하와 대기시간을 줄일 수 있는 좋은 기술 중 하나이며, 다양한 네트워크 환경에 대한 유연성과 보안성 좋아 향후 기대되는 환산 응용 기반 기술이라 할 수 있다. 본 논문에서는, 이동 에이전트 기반의 다양한 분산 응용에 효율적으로 적용할 수 있는 이동 에이전트 플랫폼인 HUMAN이 설계 구현되었다. 이동 에이전트 플랫폼, HUMAN은 파일 탐색, 그룹 어드레싱, 다양한 여정 모드, 유연 응답 기능 등의 고급 유틸리티 등을 지원하고 있어, 에이전트 프로그래밍에 있어서 높은 수월성을 제공하고 있으며, 불법파일 탐색 등과 같은 분산 응용에 특히 유용할 수 있다. 또한 에이전트 등록, 해지 등을 위한 관리 서버 등을 제공하고 있어, peer-to-peer관련 분산 응용 시스템의 개발에도 용이하게 적용할 수 있다는 장점을 가지고 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.397-399
• /
• 2001

프록시 서버는 점차 다양해지고 있는 클라이언트의 이질성(heteroginiety)과 이에 따른 서버와 네트워크의 과부하를 해결하기 위한 방법으로 제시되고 있다. 본 연구에서는 분산된 프록시 서버 환경에서 주어진 프록시들의 부하 균형을 위하여 각 비디오 객체의 계층(layer) 들을 서로 다른 프록시에 분리하여 저장하였다가 필요시 전송함으로서 네트워크 환경에서 특정 프록시 서버로의 집중과 혼잡성을 줄이도록 한다. 이로서 적응적인 화질(Quality Adaptation)을 보장하기 위한 계층적 코딩(layered coding) 스트리밍 서비스를 보다 효율적으로 제공할 수 있다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.12B
• /
• pp.811-817
• /
• 2005

Efficient bandwidth management is necessary in order to provide high quality service to users in a multimedia wireless/mobile network. In this paper, I propose an on-line load balancing algorithm with preemption. This technique is able to balance the traffic load among cells accommodating heterogeneous multimedia services while ensuring efficient bandwidth utilization. The most important features of my algorithm are its adaptability, flexibility and responsiveness to current network conditions. In addition, my online scheme to control bandwidth adaptively is a cell-oriented approach. This approach has low complexity making it practical for real cellular networks. Simulation results indicate the superior performance of my algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.2B
• /
• pp.36-43
• /
• 2005

This paper suggests a method to effectively apply an application model of fuzzy-neural network to the optimal load distribution algorithm, considering the complication and non-linearity of the web server environment. We use the clustering web server in the linux system and it consists of a load balancer that distributes the network loads and some of real servers that processes the load and responses to the client. The previous works considered only with the scrappy decision information such as the connections. That is, since the distribution algorithm depends on the input of the whole network throughput, it was proved inefficient in terms of performance improvement of the web server. With the proposed algorithm, it monitors the whole states of both network input and output. Then, it infers CPU and memory states of each real server and effectively distributes the requests of the clients. In this paper, the proposed model is compared with the previous method through simulations and we analysis the results to develop the optimal and intelligent load balancing model.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.53-58
• /
• 2013

A Fat-Tree topology has multiple paths between any pair of hosts. The delay for the multiple paths with an equal number of hops depends mainly on the queuing delay. However, most of the existing load-balancing schemes do not sufficiently exploit the characteristics of Fat-Tree. In most schemes load-balancing is performed at a flow level. Packet-level load-balancing schemes usually require the availability of special transport layer protocols to address packet reordering. In this paper, we propose a new packet-level load-balancing scheme which can enhance network utilization while minimizing packet reordering in Fat-Trees. Simulation results show that the proposed scheme provides as high TCP throughput as a randomized flow-level Valiant load balancing scheme for a best case.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.4
• /
• pp.809-816
• /
• 2010

This paper proposed about a load-balanced multi-agent model in mobile distribution environment to monitor moving patients and to deal with a situation of emergency. This model was designed to have a structure based on distribution framework by expanding a mobile system, and provides healthcare services based on real time situational information on moving patients. In order to overcome the limitation of middleware when we design system, we provided an abstract layer between applications and their base network infrastructure so that balance between QoS requests and network life can be maintained. In addition, clustering was used in cells for the efficient load distribution among multi-agents. By using Clustering FCM, we got optimal resources and had solve about transmission delay.