• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.98-106
• /
• 2013

The duty-cycle synchronization among mobile sensor nodes with energy-harvesting is very important. The nodes should keep their duty-cycle same to others as much as possible because they have to cooperate each other and to consume energy efficiently. The distribution of node position in network affects not only node connectivity but also the active time of synchronized nodes, and it relates to network life-time finally. In this paper, we introduce a network topology change algorithm (TCA) for energy-harvesting mobile sensor networks based on self-synchronized duty-cycling. The algorithm tries to change a network topology into a balanced topology where the mobile sensor nodes are unified according to the density of the number of nodes. For TCA, both fluid flow algorithm and flock dispersion algorithm are proposed and they are evaluated through the simulation in agent based modeling language. TCA is applied to the energy-harvesting mobile sensor networks to improve the synchronization of duty-cycle and to reduce the variation of energy consumption among nodes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.9B
• /
• pp.857-866
• /
• 2009

Recently, IP mobility study of infra's common elements is undergoing processes to achieve FMC based on All-IP. Existing typical mobility technology based on IP is IETF's Mobile IP. However, it faced to limitations due to packet loss, delays when MN is moving on Mobile IP, also existing network infra - routers (FA)-needed to add/change the functions to support Mobile IP. In this paper, existing mobility problems based on IP and the suggested improvements for platforms which support mobility, quality, security are proposed. It discusses the performance on the current existing IP infrastructure derived from simulation analysis on mobility packet loss and delay. From the resulting data, improvements will also be outlined for optimal performance.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.459-464
• /
• 2006

It has been well proved that shiphandling simulators are efficient and useful facilities for training and education of cadets and seafarers. Most of the maritime universities and many maritime training institutes all over the world have installed shiphandling simulators, which play important roles in maritime education and training. However, most of the Shiphandling simulators are standalone facilities with diversities on system architectures, layouts and functions. STCW78/95 requires simulators used for simulator-based training shall be suitable for the selected objectives and training tasks. To ensure the shiphandling simulator facilities meet the requirements of STCW convention and other expanded applications, collaborated research and coordination are needed in development and application of shiphandling simulators. Performance standard should be established for shiphandling simulator systems considering the advanced research needs as well as the needs in education, training, and assessment of competence. Standardizing and exchanging shiphandling mathematical models will improve critical performance of the system. Cooperated research on model course and training assessment approaches will enhance the training standard. In addition, the rapid spread of the internet technology has shown a promising future of application of shiphandling simulators through internet. Research has been carried out on internet based integration of multiple shiphandling simulators. A multi-agent based system, including necessary hardware, has been developed. Collaborated operation of the system can be of benefit in filling the gaps of the technical and operational level and methodology between maritime universities, enhancing mutual understanding of the navigation customs and culture background among cadets and seafarers from different countries, facilitating communication and maritime English training, and extending the functions of shiphandling simulators.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1153-1160
• /
• 2021

This paper proposes a routing algorithm that determines the optimal path using deep reinforcement learning in software-defined networks. The deep reinforcement learning model for learning is based on DQN, the inputs are the current network state, source, and destination nodes, and the output returns a list of routes from source to destination. The routing task is defined as a discrete control problem, and the quality of service parameters for routing consider delay, bandwidth, and loss rate. The routing agent classifies the appropriate service class according to the user's quality of service profile, and converts the service class that can be provided for each link from the current network state collected from the SDN. Based on this converted information, it learns to select a route that satisfies the required service level from the source to the destination. The simulation results indicated that if the proposed algorithm proceeds with a certain episode, the correct path is selected and the learning is successfully performed.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.55-62
• /
• 2023

In this paper, we present a method that utilizes geometry-based sound generation techniques to efficiently handle multiple sound sources, wind turbulence, and temperature-dependent interactions. Recently, a method based on reduced raytracing has been proposed to update the sound position and efficiently calculate sound propagation and diffraction without recursive reflection/refraction of many rays, but this approach only considers the propagation characteristics of sound and does not consider the interaction of multiple sound sources, wind currents, and temperature. These limitations make it difficult to create sound scenes in a variety of virtual environments because they only generate static sounds. In this paper, we propose a method for efficiently constructing a sound map in a situation where multiple sounds are placed, and a method for efficiently controlling the movement of an agent through it. In addition, we propose a method for controlling sound propagation by considering wind currents and temperature. The method proposed in this paper can be utilized in various fields such as metaverse environment design and crowd simulation, as well as games that can improve content immersion based on sound.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.9
• /
• pp.407-418
• /
• 2023

The Multi-Object Goal Visual Navigation(MultiOn) is a visual navigation task in which an agent must visit to multiple object goals in an unknown indoor environment in a given order. Existing models for the MultiOn task suffer from the limitation that they cannot utilize an integrated view of multimodal context because use only a unimodal context map. To overcome this limitation, in this paper, we propose a novel deep neural network-based agent model for MultiOn task. The proposed model, MCFMO, uses a multimodal context map, containing visual appearance features, semantic features of environmental objects, and goal object features. Moreover, the proposed model effectively fuses these three heterogeneous features into a global multimodal context map by using a point-wise convolutional neural network module. Lastly, the proposed model adopts an auxiliary task learning module to predict the observation status, goal direction and the goal distance, which can guide to learn the navigational policy efficiently. Conducting various quantitative and qualitative experiments using the Habitat-Matterport3D simulation environment and scene dataset, we demonstrate the superiority of the proposed model.

• Journal of Korea Multimedia Society
• /
• v.2 no.3
• /
• pp.339-346
• /
• 1999

When a multicast video conference system utilizes RTP (Real Time Protocol) and RTCP (Real Time Control Protocol), the loss rate and the synchronization of transfer in RTCP affect the scalability of the system. The random delay technique introduced to resolve the problems is so simple that leads the network to meet some congestion in synchronizing feedback information when lots of people transfer the feedback information simultaneously, which reduces the scalability of system. In this paper, we propose a new feedback control algorithm that provides priority levels with the RTCP packet, which cuts down the feedback delay and increases the scalability. The criteria of providing priority Based on the decided priority level, Agent forced the session participants to provide much more RTCP packets, positively controlled, and the possible bandwidth can be measured. The simulation on this technique decreases the delay, and the feedback messages are equally distributed on a given time period.

• Journal of the Korean Society of Safety
• /
• v.32 no.4
• /
• pp.16-21
• /
• 2017

The sprinkler system is a basic fire extinguishing system that uses water as an extinguishing agent. In order to evaluate the fire extinguishing performance of the sprinkler system, information such as the discharge angle, discharge speed, discharge pressure, flow rate, and water droplet size of the installed head are required. However, there is a lack of research on droplets size compared to other requirements. In this study, to evaluate the extinguishing characteristics of sprinkler system, the droplet size distribution was measured for various types of sprinkler heads actually used. The size of the droplet was measured using laser diffraction method. The 50% cumulative volume distribution ($D_{v50}$) according to discharge coefficient(K factor) was $540{\mu}m{\sim}695{\mu}m$ for K50, $542{\mu}m{\sim}1,192{\mu}m$ for K80, $980{\mu}m{\sim}1,223{\mu}m$ for K115 and $1,188{\mu}m{\sim}1,234{\mu}m$ for K202. Based on the measured results, the vaeiance of the droplet particle distribution and the distribution ($D_{v50}$) according to discharge coefficient(K factor) was $540{\mu}m{\sim}695{\mu}m$ for K50, $542{\mu}m{\sim}1,192{\mu}m$ for K80, $980{\mu}m{\sim}1,223{\mu}m$ for K115 and $1,188{\mu}m{\sim}1,234{\mu}m$ for K202. Based on the measured results, the vaeiance of the droplet particle distribution and the Rosin-Rammler index value are presented. As a result of the fire simulation with FDS, it was confirmed that the performance difference occurs according to the water droplet size distribution even when the same amount of water is used. Therefore, the extinguishing performance of the sprinkler system should be evaluated considering the droplet size distribution according to the sprinkler head type.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.517-521
• /
• 2008

A worm is a malware that propagates quickly from host to host without any human intervention. Need of early worm detection has changed research paradigm from signature based worm detection to the behavioral based detection. To increase effectiveness of proposed solution, in this paper we present mechanism of detection and prevention of worm in distributed fashion. Furthermore, to minimize the worm destruction; upon worm detection we propagate the possible attack aleγt to neighboring nodes in secure and organized manner. Considering worm behavior, our proposed mechanism detects worm cycles and infection chains to detect the sudden change in network performance. And our model neither needs to maintain a huge database of signatures nor needs to have too much computing power, that is why it is very light and simple. So, our proposed scheme is suitable for the ubiquitous environment. Simulation results illustrate better detection and prevention which leads to the reduction of infection rate.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.10B
• /
• pp.630-637
• /
• 2007

In this paper, we propose an effective handoff mechanism using events at the layer 2 and a tunnel-based mechanism that is performed by sharing information of a mobile node and anticipated-based handoff. Also, we propose a novel anticipated-reservation mechanism using the events triggered at layer 2 to make easy a resource reservation procedure. Comparing with the existing mobile IPv6 handoff mechanism, our proposed handoff mechanism can reduce an amount of handoff latency and traffic volume between access routers where handoff is performed. In the anticipated-reservation mechanism, the amount of signaling traffics between an mobile node and the correspondent nodes can be reduced and then the allocation of resource can be accomplished. We present the performance of our proposed handoff mechanism in the wireless network environment by ns-2 simulation. Although the anticipated-reservation mechanism including a proxy agent is explained without performance evaluation, we can validate the operation of our proposed model only by the procedure of message flow.