• 대한교통학회지
• /
• 제19권6호
• /
• pp.183-194
• /
• 2001

중앙집중형 도로교통정보시스템은 실시간 교통정보를 수집하고, 사용자의 요청을 받아 경로탐색, 위치정보, 목적지탐색 등의 정보를 전달해주는 시스템이다. 이러한 시스템에서 서버는 매우 많은 클라이언트로부터 경로탐색 요청을 받게 되며, 이 요청을 서버에서 효율적으로 처리해야 하는 다중경로탐색 알고리즘이 필요하다. 본 연구에서는 다중경로탐색을 수행하기 위하여, 주기적으로 연산된 이웃노드의 최단경로탐색 결과를 이용하여 클라이언트의 경로탐색 수행시간을 감소시키는 휴리스틱(Heuristic) 알고리즘을 제시한다. 본 연구에서 제시하는 이웃노드 최단경로탐색 결과를 이용한 다중경로탐색 알고리즘은 많은 경우에 최단과 동일한 결과를 나타내며, 최단이 아닌 경우에도 최단경로 값과 오차가 크지 않으면서도 연산시간을 많이 줄일 수 있는 알고리즘이고, 도로교통과 같은 토폴로지(Topology) 형태에 효과적으로 적용되고, 계층을 이루는 형태의 모델에서도 효율적인 결과를 나타낸다. 이웃노드 최단경로탐색 결과를 이용한 다중경로탐색 알고리즘의 경로탐색시간은 다른 꼬리표설정 알고리즘과 꼬리표개선 알고리즘보다 50배 이상 빨랐으며, 경로탐색 결과가 최단이 아닌 경우 0.1%이하의 거리오차가 발생했다.

• 한국소음진동공학회논문집
• /
• 제27권2호
• /
• pp.195-203
• /
• 2017

This paper is concerned with the active vibration control of shell structure that is subjected to internal unbalanced excitation by using active mounts and accelerometers. The unbalanced excitation is caused by a rotating unbalanced mass. The control algorithm considered in this study is the negative acceleration feedback (NAF) control. A simplified dynamic model was derived to verify the effectiveness of the NAF control. Four actuators and four accelerometers were mounted on the shell structure, so that the multiple-input and multiple-output (MIMO) NAF controller was designed by both centralized and decentralized ways. Numerical results show that both the decentralized and centralized NAF controllers are effective. Based on the numerical simulation, the proposed decentralized NAF controller was applied to the real shell structure. Experimental results show that the proposed decentralized NAF controller can effectively suppress vibrations of the shell structure.

• Journal of Information Processing Systems
• /
• 제13권6호
• /
• pp.1613-1627
• /
• 2017

The concept of the Internet of Things (IoT) enables physical objects or things to be virtually accessible for both consuming and providing services. Undue access from irresponsible activities becomes an interesting issue to address. Maintenance of data integrity and privacy of objects is important from the perspective of security. Privacy can be achieved through various techniques: password authentication, cryptography, and the use of mathematical models to assess the level of security of other objects. Individual methods like these are less effective in increasing the security aspect. Comprehensive security schemes such as the use of frameworks are considered better, regardless of the framework model used, whether centralized, semi-centralized, or distributed ones. In this paper, we propose a new semi-centralized security framework that aims to improve privacy in IoT using the parameters of trust and reputation. A new algorithm to elect a reputation coordinator, i.e., ConTrust Manager is proposed in this framework. This framework allows each object to determine other objects that are considered trusted before the communication process is implemented. Evaluation of the proposed framework was done through simulation, which shows that the framework can be used as an alternative solution for improving security in the IoT.

• 전기학회논문지
• /
• 제60권3호
• /
• pp.566-571
• /
• 2011

As Plug-in Hybrid Vehicle and Electric Vehicle (PHEV/EV) take a greater share in the personal automobile market, their high penetration levels may bring potential challenges to electric utility especially at the distribution level. Thus, there is a need for the flexible charging management strategy to compromise the benefits of both PHEV/EV owners and power grid side. There are many different management methods that depend on the objective function and the constraints caused by the system. In this paper, the schema and dispatching schedule of centralized PHEV/EV charging spot network are analyzed. Also, we proposed and compared three power allocation strategies for centralized charging spot. The first strategy aims to maximize state of vehicles at plug-out time, the rest methods are equalized allocation and prioritized allocation based on vehicles SoC. The simulation results show that each run of the optimized algorithms can produce the satisfactory solutions to response properly the requirement from PHEV/EV customers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권3호
• /
• pp.958-979
• /
• 2023

Due to laws, regulations, privacy, etc., between 70-90 percent of providers do not share medical data, forming a "data island". It is essential to collaborate across multiple institutions without sharing patient data. Most existing methods adopt distributed learning and centralized federal architecture to solve this problem, but there are problems of resource heterogeneity and data heterogeneity in the practical application process. This paper proposes a collaborative deep learning modelling method based on the blockchain network. The training process uses encryption parameters to replace the original remote source data transmission to protect privacy. Hyperledger Fabric blockchain is adopted to realize that the parties are not restricted by the third-party authoritative verification end. To a certain extent, the distrust and single point of failure caused by the centralized system are avoided. The aggregation algorithm uses the FedProx algorithm to solve the problem of device heterogeneity and data heterogeneity. The experiments show that the maximum improvement of segmentation accuracy in the collaborative training mode proposed in this paper is 11.179% compared to local training. In the sequential training mode, the average accuracy improvement is greater than 7%. In the parallel training mode, the average accuracy improvement is greater than 8%. The experimental results show that the model proposed in this paper can solve the current problem of centralized modelling of multicenter data. In particular, it provides ideas to solve privacy protection and break "data silos", and protects all data.

• Journal of information and communication convergence engineering
• /
• 제5권2호
• /
• pp.93-97
• /
• 2007

In this paper, centralized access control and slot allocation algorithm is proposed for wireless networks. The proposed algorithm is characterized by the contention-based reservation. In order to reduce the collision probability of reservation request, the base station calculates and broadcasts the transmission probability of reservation requests, and the wireless terminal transmits its reservation request with the received transmission probability. The scheduler allocates the uplink data slots based on the successful reservation requests. Simulation results show that the proposed algorithms can provide high channel utilization, and furthermore, maintains constant delay performance in the heavy traffic environment.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2008년도 추계학술대회 및 정기총회
• /
• pp.494-497
• /
• 2008

In Mobile ad hoc networks (MANET), nodes could transmit packets for each other to communicate beyond their transmission range without centralized control. Unlike infrastructure-based wireless networks, due to the unique characteristics of MANETs such as open network architecture, stringent resource constraints and highly dynamic network topology, networks are vulnerable to wormhole attacks launched through colluding nodes. In this paper, we develop an wormhole detection and prevention algorithm for MANET.

• Smart Structures and Systems
• /
• 제22권5호
• /
• pp.547-560
• /
• 2018

This paper investigates the operation of the $H_{\infty}$ static output-feedback controller to reduce dynamic responses under seismic excitation on the five-story and benchmark 20 story building with parametric uncertainties. Linear matrix inequality (LMI) control theory is applied in this system and then to achieve the desired LMI formulations, some transformations of the LMI variables is used. Conversely uncertainties due to material properties, environmental loads such as earthquake and wind hazards make the uncertain system. This problem and its effects are studied in this research. Also to decrease the transition of large amount of data between sensors and controller, avoiding the disruption of whole control system and economy problems, the operation of the decentralized controllers is investigated in this paper. For this purpose the comparison between the performance of the centralized, fully decentralized and partial decentralized controllers in uncoupled and coupled cases is performed. Also, the effect of the changing the number of stories in substructures is considered. Based on the numerical results, the used control algorithm is very robust against the parametric uncertainties and structural responses are decreased considerably in all the control cases but partial decentralized controller in coupled form gets the closest results to the centralized case. The results indicate the high applicability of the used control algorithm in the tall shear buildings to reduce the structural responses and its robustness against the uncertainties.

• 한국통신학회논문지
• /
• 제24권7B호
• /
• pp.1229-1238
• /
• 1999

무선 ATM의 매체 접근 제어 프로토콜에서의 통계적 다중화를 실현하기 위해서는 중앙의 무선 ATM 액세스점(Wireless ATM Access Point)에 의해 동적 슬롯 할당을 수행할 수 있는 스케쥴링 기능이 제공되어야 한다. 그러므로 AP에서는 이동국으로부터 전송된 동적 파라미터(Dynamic Parameter)를 이용해 중앙 집중식 스케쥴링 기법을 적용하는 동적 대역 할당 매체 접근 제어 프로토콜이 요구된다. 본 논문에서는 묵시적인 방법에 의한 간접적인 동적 파라미터의 전송을 통해 동적 슬롯 할당이 가능한 새로운 스케쥴링 알고리즘을 제안한다. 제안된 방식을 알고리즘의 적시성과 정확성 측면에서 우수한 성능을 보이는 것을 확인함으로써 동적 파라미터에 전적으로 의존하지 않고 효율적인 스케쥴링이 수행될 수 있음을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.2936-2951
• /
• 2016

An emerging satellite technology, Operationally Responsive Space (ORS) is expected to provide a fast and flexible solution for emergency response, such as target tracking, dense earth observation, communicate relaying and so on. To realize large distance transmission, we propose the use of available relay satellites as relay nodes. Accordingly, we apply software defined network (SDN) technology to ORS networks. We additionally propose a satellite network architecture refered to as the SDN-based ORS-Satellite (Sat) networking scheme (SDOS). To overcome the issures of node failures and dynamic topology changes of satellite networks, we combine centralized and distributed routing mechanisms and propose a fast recovery routing algorithm (FRA) for SDOS. In this routing method, we use centralized routing as the base mode.The distributed opportunistic routing starts when node failures or congestion occur. The performance of the proposed routing method was validated through extensive computer simulations.The results demonstrate that the method is effective in terms of resoving low end-to-end delay, jitter and packet drops.