• Annual Conference of KIPS
• /
• 2014.04a
• /
• pp.319-321
• /
• 2014

카카오톡, 페이스북 등의 소셜 네트워크 서비스(Social Network Service: SNS)가 발전함에 따라 이를 기반으로 하는 소셜 네트워크 게임(Social Network Game: SNG)이 지인과 함께 가볍게 즐길 수 있다는 장점을 내세워 유저들로부터 많은 호응을 얻고 있다. 특히 SNG가 갖는 장점은 언제 어디서나 즐길 수 있는 스마트폰 앱에 접목되면서 그 효과가 극대화되고 있다. 이에 많은 개발사들이 앞다투어 SNG 개발에 참여를 하고 있는 상황이지만 다양한 SNS 서비스 플랫폼에 비해 SNG 개발을 위한 공개 엔진은 없는 상황이다. 본 논문에서는 다양한 SNS 플랫폼을 통해 공통으로 사용할 수 있는 통합 관리 소셜 네트워크 엔진을 개발하는 과정과 구현된 엔진을 사용하여 디바이스를 통한 그래픽 출력 성능을 실험하였으며, 실험 결과 기존의 SNG와의 속도차이가 없음을 보여 주어 게임 구현하기에 충분함을 입증하였다.

• Journal of the Korea Safety Management & Science
• /
• v.2 no.2
• /
• pp.177-185
• /
• 2000

In recent dates, Internet advertising effects are expanded by the steep increment of the Internet users and the extension of the advertising market will be accelerated through Internet. This paper indicates the importance of Internet advertising and suggests the solution of a network advertising service. The system is divided into an Advertise Server, an Advertiser and a Web Publisher. This study proposes both the collection and the analysis of traffic data in real time. Also, the banner advertising frames are smoothed for the impression using the solution for the banner exchange engine and are developed using various impression methods, that is, Fixed/variable Banner, Scheduling Banner, Multi-Impression Banner, and Frame Remote Control to increase the advertising effect. And then to increase the advertising effect web publisher, which is specialized in lottery site, is constructed using the network advertising service and various advertising technology.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.505-517
• /
• 2021

This research presented an implementation of autostereoscopic virtual three-dimensional (3D) button device as non-contact style. The proposed device has several characteristics about visible feature, non-contact use and artificial intelligence (AI) engine. The device was designed to be contactless to prevent virus contamination and consists of 3D buttons in a virtual stereoscopic view. To specify the button pressed virtually by fingertip pointing, a simple deep learning network having two stages without convolution filters was designed. As confirmed in the experiment, if the input data composition is clearly designed, the deep learning network does not need to be configured so complexly. As the results of testing and evaluation by the certification institute, the proposed button device shows high reliability and stability.

• The Journal of Korea Robotics Society
• /
• v.17 no.2
• /
• pp.118-123
• /
• 2022

We implemented a real-time cloud robotics application by offloading robot navigation engine over to 5G Mobile Edge Computing (MEC) sever. We also ran a fleet management system (FMS) in the server and controlled the movements of multiple robots at the same time. The mobile robots under the test were connected to the server through 5G SA network. Public 5G network, which is already commercialized, has been temporarily modified to support this validation by the network operator. Robot engines are containerized based on micro-service architecture and have been deployed using Kubernetes - a container orchestration tool. We successfully demonstrated that mobile robots are able to avoid obstacles in real-time when the engines are remotely running in 5G MEC server. Test results are compared with 5G Public Cloud and 4G (LTE) Public Cloud as well.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.9
• /
• pp.116-123
• /
• 1997

We have already presented the constraint satisfaction algorithm which could solve the closed loop porblem in constraint network by using local constraint propagation, variable elimination and constraint modularization. With this algorithm, we have implemented a knowledge-based system (intelligent CAD) for supporting machine design interactively. In this paper, we present newer constraint satisfaction algorithm which can solve inequalities or under-constrained problems in constraint network, interactively and effi- ciently. This algorithm is a hybrid type of using both declarative description (constraint representation) and optimization algorithm (Simulated Annealing), simultaneously. The under-constrained problems are represented by constraint networks and satisfied completely with this algorithm. The usefulness of our algorithm will be illustrated by the application to a gear design.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2006.11a
• /
• pp.281-284
• /
• 2006

An intelligent performance diagnostic program using the Neural Network was proposed for PW206C turboshaft engine. It was selected as a power plant for the tilt rotor type Smart UAV (Unmanned Aerial Vehicle) which has been developed by KARI (Korea Aerospace Research Institute). The measurement parameters of Smart UAV propulsion system are gas generator rotational speed, power turbine rotational speed, exhaust gas temperature and torque. But two measurement such as compressor exit pressure and compressor turbine exit temperature were added because they were difficult each component diagnostics using the default measurement parameter. The performance parameters for the estimate of component performance degradation degree are flow capacities and efficiencies for compressor, compressor turbine and power turbine. Database for network learning and test was constructed using a gas turbine performance simulation program. From application results for diagnostics of the PW206C turboshaft engine using the learned networks, it was confirmed that the proposed diagnostics could detect well the single fault types such as compressor fouling and compressor turbine erosion.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.5
• /
• pp.1008-1013
• /
• 2009

In this paper, we proposed the biological signal (body heat, pulse, breathe rate, and blood pressure) analysis system using wireless sensor. In order to analyze, we designed a back-propagation neural network system using expert group system. The proposed system is consist of hardware patt such as UStar-2400 ISP and Wireless sensor and software part such as Knowledge Base module, Inference Engine module and User Interface module which is inserted in Host PC. To improve the accuracy of the system, we implement a FEC (Forward Error Correction) block. For conducting simulation, we chose 100 data sets from Knowledge Base module to train the neural network. As a result, we obtained about 95% accuracy using 128 data sets from Knowledge Base module and acquired about 85% accuracy which experiments 13 students using wireless sensor.

• 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.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.10
• /
• pp.323-328
• /
• 2020

Recently, new approaches that significantly improve performance in object detection and recognition using deep learning technology have been proposed quickly. Of the various techniques for object detection, especially facial object detection (Faster R-CNN, R-CNN, YOLO, SSD, etc), SSD is superior in accuracy and speed to other techniques. At the same time, multiple object detection networks are also readily available. In this paper, among object detection networks, Mobilenet v2 network is used, models combined with SSDs are trained, and methods for detecting objects at a rate of four times or more than conventional performance are proposed using TensorRT engine, and the performance is verified through experiments. Facial object detector was created as an application to verify the performance of the proposed method, and its behavior and performance were tested in various situations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.4B
• /
• pp.368-376
• /
• 2009

Over recent years, the development of multimedia devices has meant that a wider multimedia streaming service can be supported, and there are now many ways in which TV channels can communicate with different terminals. Generally, scalable video streaming is known to provide more efficient channel capacity than simulcast video streaming. Simulcast video streaming requires a large network bandwidth for all resolutions, but scalable video streaming needs only one flow for all resolutions. On the contrary, to preserve the same video quality, SVC(Sealable Video Coding) needs a higher bit-rate than AVC(non-layered Video Coding) due to the coding penalty($10%{\sim}30%$). In previous research, scalable video streaming has been compared with simulcast video streaming for network channel capacity, in two-user simulation environments. The simulation results show that the channel capacity of SVC is $16{\sim}20%$ smaller than AVC, but scalable video streaming is not efficient because of the limit of the present network framework. In this paper, we propose a new network framework with a new router using EDE(Extraction Decision Engine) and SVC Extractor to improve network performance. In addition, we compare the SVC environment in the proposed framework with previous research on the same way subject. The proposed network framework shows a channel capacity 50%(maximum) lower than that found in previous research studies.