• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.12 no.2
• /
• pp.38-44
• /
• 2013

In this paper, we designed and implemented a mobile video transmission device with 1080i resolution and 30fps frame rate using wireless multi-access technologies. The video acquisition delay time of 16us and jitter of 21ms for this equipment were obtained, respectively. The device is designed to access various communication networks. In order to evaluate the performance of its video transmission capability, a server and monitoring viewer are implemented. In the performance test using LTE network, 211ms delay time was obtained at 3Mbps transmission. Rendering of 1fps frame rate at the 265kbps transmission was confirmed for the transmission test using Inmarsat network.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.8
• /
• pp.1049-1062
• /
• 2002

As the distributed resources in the networks have become increasingly popular, the accesses to these resources having been activated. Especially, the accesses to the distributed resources using the mobile agent technologies provide the mechanisms supporting mobility with mobile users as well as the dynamic accesses to the resources in the fixed networks. Proxy-based mobile agent model is defined as mobile agent network model which allocates the hierarchical domains to the distributed resources changed dynamically, assigns one proxy server for each domain, and promotes the management and the cooperation of the mobile agents. In this paper, we introduce the architecture and the execution scenario for proxy-based mobile agent model which is suitable for the hierarchical management domains. In simulation, we evaluate the proxy server's route optimization functionality and the performance reducing execution time of the mobile agents.

• Journal of Digital Convergence
• /
• v.15 no.3
• /
• pp.201-209
• /
• 2017

This paper is a smartphone object recognition system using hierarchical convolutional neural network. The overall configuration is a method of communicating object information to the smartphone by matching the collected data by connecting the smartphone and the server and recognizing the object to the convergence neural network in the server. It is also compared to a hierarchical convolutional neural network and a fractional convolutional neural network. Hierarchical convolutional neural networks have 88% accuracy, fractional convolutional neural networks have 73% accuracy and 15%p performance improvement. Based on this, it shows possibility of expansion of T-Commerce market connected with smartphone and broadcasting media.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2003.04a
• /
• pp.37-44
• /
• 2003

The aim of this research was performance of Web-based real time DGPS using MS visual basic 6.0 Active X control. This system supported the navigation of rover by transmitting the DGPS correction information of base station via internet and monitored the moving of rover by displaying the corrected results received from rover as well as permitting Web users monitoring for rover. The Web-based DGPS server and the client system using MS visual basic 6.0 Active X control for DGPS displayed on Web-browser with the pre analysis was constructed and the system loading test via Web-browser was performed after debugging this system through the static GPS test and publishing it on Web. Also, the wireless internet DGPS test on Web by changing the distance between a base station and a rover was carried out. As the result of the test, the real time DGPS could be performed by transmitting the correction message which calculated from the server system to the client system continually. The rover could be managed by displaying the location information which received from the client system. Other users also could monitor the rover by connecting with this system via internet in real time.

• The KIPS Transactions:PartC
• /
• v.13C no.7 s.110
• /
• pp.923-932
• /
• 2006

For the successful promotion about the NGN (Next Generation Network), centralized network control and management technology is required to guarantee QoS and Traffic Engineering. This paper describes SNMP. Web Services and Network Open Interface based connection management scheme which can be used in centralized server based network in which the control plane is separated from transport plane. We also constructed platforms about each of these connection management schemes and performed comparing and analyzing the performance of connection setup delays.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.10
• /
• pp.556-568
• /
• 2003

This paper presents and analyzes the performance of a weighted-window based cache replacement policy for the efficient media caching in streaming media cache servers. The proposed policy makes, for each cached object, use of the reference count, reference time, amount of media delivered to clients and, in particular, the periodic patterns of user requests. Also, by giving weights to the recently referenced media contents, the replacement policy adequately and swiftly reflects the ever-changing characteristics of users preferences. The simulation studies show that the performance of the proposed policy has improved over the conventional policies such as LRU, LFU and SEG - in terms of hit ratio, byte hit ratio, delayed start and cache input.

• Journal of Advanced Navigation Technology
• /
• v.6 no.4
• /
• pp.290-295
• /
• 2002

In Mpeg-4 system, objects are composed of ES(Elementary Stream). Each of the objects is managed independently by object-based coding and is transmitted via DMIF(Delivery Multimedia Integration Framework). The data streams in Mpeg-4 are transmitted with using DVB-Data Carousel which can improve the reliability and efficiency by cyclic retransmission. This paper describes a system which transmits some part of data stream of Mpeg-4 object with using DVB-Data Carousel to clients via satellite. This also uses a performance enhancing proxy server for reducing round trip time between ground network and satellite.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.5
• /
• pp.926-932
• /
• 2007

Various research groups proposed various architecture of hardware VPN for the high performance VPN system. However, the VPN based on hardware researcher are focused only on the encryption acceleration. Soft based VPN is only useful when the network connection is slow. We have to consider the hardware performance (encryption/decryption processing capability, packet processing, architecture method) to implement hardware based VPN. In this paper, we have analysed architecture of hardware, consideration and problems for high-speed VPN system, From the result, we can choose the proper design guideline.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.911-931
• /
• 2021

Recently, most cloud services use Docker container environment to provide their services. However, there are no researches to evaluate the performance of communication libraries for multi-GPU based distributed deep learning in a Docker container environment. In this paper, we propose an efficient communication architecture for multi-GPU based deep learning in a Docker container environment by evaluating the performances of various communication libraries. We compare the performances of the parameter server architecture and the All-reduce architecture, which are typical distributed deep learning architectures. Further, we analyze the performances of two separate multi-GPU resource allocation policies - allocating a single GPU to each Docker container and allocating multiple GPUs to each Docker container. We also experiment with the scalability of collective communication by increasing the number of GPUs from one to four. Through experiments, we compare OpenMPI and MPICH, which are representative open source MPI libraries, and NCCL, which is NVIDIA's collective communication library for the multi-GPU setting. In the parameter server architecture, we show that using CUDA-aware OpenMPI with multi-GPU per Docker container environment reduces communication latency by up to 75%. Also, we show that using NCCL in All-reduce architecture reduces communication latency by up to 93% compared to other libraries.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.54-60
• /
• 2021

AR (Augmented Reality) is a technology that provides virtual content to the real world and provides additional information to objects in real-time through 3D content. In the past, a high-performance device was required to experience AR, but it was possible to implement AR more easily by improving mobile performance and mounting various sensors such as ToF (Time-of-Flight). Also, the importance of mobile augmented reality is growing with the commercialization of high-speed wireless Internet such as 5G. Thus, this paper proposes a system that can provide AR services via GNN (Graph Neural Network) using cameras and sensors on mobile devices. ToF of mobile devices is used to capture depth maps. A 3D point cloud was created using RGB images to distinguish specific colors of objects. Point clouds created with RGB images and Depth Map perform downsampling for smooth communication between mobile and server. Point clouds sent to the server are used for 3D object detection. The detection process determines the class of objects and uses one point in the 3D bounding box as an anchor point. AR contents are provided through app and web through class and anchor of the detected object.