• The KIPS Transactions:PartC
• /
• v.13C no.6 s.109
• /
• pp.767-774
• /
• 2006

In order to support overlay multicast with scalability and reliability, in this paper, we introduced a new multicast tree construction method and a regional multicast. The architecture we introduce efficiently transmits streaming data to the users from multicast source in an environment in which non-multicast network and multicast network. The new multicast tree construction method enables the connection with a remote relay which is an optimum condition in case there is no relay in subnet. Besides, by defining a new service type to traffic class of IPv6 header, a selective adoption or FEC and a distinctive transmission became possible. In order to verify the effect of the solution we suggested, we embodied an environment which is similar to the actual service environment by a simulator and tested it. The result of the test shows that the proposed method is more efficient than the traditional unicast approach method and the existing overlay multicast.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3567-3582
• /
• 2020

Human activities are often affected by weather conditions. Automatic weather recognition is meaningful to traffic alerting, driving assistance, and intelligent traffic. With the boost of deep learning and AI, deep convolutional neural networks (CNN) are utilized to identify weather situations. In this paper, a three-channel convolutional neural network (3C-CNN) model is proposed on the basis of ResNet50.The model extracts global weather features from the whole image through the ResNet50 branch, and extracts the sky and ground features from the top and bottom regions by two CNN5 branches. Then the global features and the local features are merged by the Concat function. Finally, the weather image is classified by Softmax classifier and the identification result is output. In addition, a medium-scale dataset containing 6,185 outdoor weather images named WeatherDataset-6 is established. 3C-CNN is used to train and test both on the Two-class Weather Images and WeatherDataset-6. The experimental results show that 3C-CNN achieves best on both datasets, with the average recognition accuracy up to 94.35% and 95.81% respectively, which is superior to other classic convolutional neural networks such as AlexNet, VGG16, and ResNet50. It is prospected that our method can also work well for images taken at night with further improvement.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.3
• /
• pp.85-93
• /
• 2012

In this paper, I designed the satellite system, using the MF-TDMA (Multi Frequency-Time Division Multiple Access) DAMA (Demand-Assigned Multiple Access) method, that allocates the satellite resources more effectively to prevent a large quantity data discard when Earth Terminals (ETs) request their satellite network resource to Network Controller (NC) by using their history information, such as traffic pattern or traffic class which have been receiving, and the network controller allocates the satellite network resource and send to earth terminals by using his history information that earth terminals have been requested already. The U.S. Military Warfighter Information Network-Tactical (WIN-T) community has selected the L-3 Linkabit MF-TDMA DAMA Network Centric Waveform (NCW) as the networking standard for full-mesh IP over SHF satellite communications. In the MF-TDMA DAMA satellite network, network controller allocates the satellite network resources and enables maximum 255 earth terminals to communicate each other for periodic satellite network resource requests of earth terminals.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2003.05a
• /
• pp.8-13
• /
• 2003

Historically, IP-based internets have been able to provide a simple best-effort delivery service to all applications they carry. Best effort treats all packets equally, with no service level, packet loss, and delay. But the needs of users have changed. The want to use the new real-time, multimedia, and multicasting applications. Thus, there is a strong need to be able to support a variety of traffic with a variety of quality-of-service requirements. The DiffServ architecture, proposed by the Internet Engineering Task Force(IETF), has become the most viable solution for provising QoS over IP networks. The DiffServ architecture does not specify any handling method between AF out-profile packets and BE packets. This paper propose a mechanism for supporting inter class fairness in the DiffServ architecture. Ⅰ proposed a modified Weighted Round Robin method to protect the BE traffic from AF out-profile packets in the core routers. The proposed technique is evaluated through simulation. Simulation results indicate that the proposed method provides better protection not only for BE packets from AF out-profile packets, but also for the AF in-profile packets in congested networks.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.11-20
• /
• 2013

In designing Intelligent Traffic Systems, it should be necessary to consider telecommunications, appearance, environment, auxiliary functions, safety, and so on. Also, in choosing a car, a consumer considers those properties. This paper tried to elucidate the fact that car color has a very significant meaning for car safety when administrating intelligent traffic services and making car-purchasing decision. We first studied on occurrence probability of car accident according to car color that has something to do with car safety. Then, we studied on the concepts of advancing color and receding color. Advancing color causes less accidents since the color looks closer than it actually is. And receding color causes more accidents since the color looks farther than it actually is. And we classified car colors into eight classes and assign their ranking to each class, considering the number of car accidents. We tried to verify our research by use of telephone questionnaire for residents in Kunsan, Republic of Korea.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.1
• /
• pp.85-94
• /
• 2002

EMRCA(Explicit Max_min Rate Control Algorithm) switch, which has been proposed in the ATM(Asychronous Transfer Mode) standard, controls the ABR(Available Bit Rate) service traffic in the ATM networks. The ABR service class of ATM networks uses a feedback control mechanism to adapt to varying link capacities. The VS/VD(Virtual Source/Virtual Destination) technique offers the possibility to segment the otherwise end-to-end ABR control loop into separate loops. The improved feedback delay and the control of ABR traffic inside closed segments provide a better performance and QoS(Quality of Service) for ABR connections with respect to throughput, delay, and jitter. This paper is study of an ABR VS/VD flow control method. Linear control theory offers the means to derive correct choices of parameters and to assess performance issues, like stability of the system, during the design phase. The performance goals are a high link utilization, fair bandwidth distribution and robust operation in various environments, which are verified by discrete event simulations. The major contribution of this work is the use of linear control theory to model and design an ABR flow control method tailored for the special layout of a VS/VD switch, the simulation shows that this techniques better than conventional method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.6
• /
• pp.1087-1094
• /
• 2008

Recently, XML has been widely used as a standard for a data exchange, and there has emerged the tendency that the size of XML document becomes larger. The data transfer can cause problems due to the increase in traffic, especially when a massive data such as Data Warehouse is being collected and analyzed. Therefore, an XMDR wrapper can solve this problem since it analyzes the tree structures of XML Schema, regenerates XML Schema using the analyzed tree structures, and sends it to each station with an XMDR Query. XML documents which are returned as an outcome encode XML tags according to XML Schema, and send standardized messages. As the formalized XML documents decrease network traffic and comprise XML class information, they are efficient for extraction, conversion, and alignment of data. In addition, they are efficient for the conversion process through XSLT, too, as they have standardized forms. In this paper we profuse a method in which XML Schema and XMDR_Query sent to each station are generated through XMDR(extended Meta-Data Registry) and the generation of products and XML conversion occur in each station wrapper.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.11
• /
• pp.5568-5587
• /
• 2018

This research uses artificial intelligence methods for computer network intrusion detection system modeling. Primary classification is done using self-organized maps (SOM) in two levels, while the secondary classification of ambiguous data is done using Sugeno type Fuzzy Inference System (FIS). FIS is created by using Adaptive Neuro-Fuzzy Inference System (ANFIS). The main challenge for this system was to successfully detect attacks that are either unknown or that are represented by very small percentage of samples in training dataset. Improved algorithm for SOMs in second layer and for the FIS creation is developed for this purpose. Number of clusters in the second SOM layer is optimized by using our improved algorithm to minimize amount of ambiguous data forwarded to FIS. FIS is created using ANFIS that was built on ambiguous training dataset clustered by another SOM (which size is determined dynamically). Proposed hybrid model is created and tested using NSL KDD dataset. For our research, NSL KDD is especially interesting in terms of class distribution (overlapping). Objectives of this research were: to successfully detect intrusions represented in data with small percentage of the total traffic during early detection stages, to successfully deal with overlapping data (separate ambiguous data), to maximize detection rate (DR) and minimize false alarm rate (FAR). Proposed hybrid model with test data achieved acceptable DR value 0.8883 and FAR value 0.2415. The objectives were successfully achieved as it is presented (compared with the similar researches on NSL KDD dataset). Proposed model can be used not only in further research related to this domain, but also in other research areas.

• Journal of Advanced Navigation Technology
• /
• v.22 no.2
• /
• pp.111-122
• /
• 2018

At present, the ground transportation system is saturated in many countries including Korea. To overcome this problem, many researches of developing a roadable personal air vehicle (PAV) are being carried out to alleviate traffic congestion and to accomplish door-to-door mobility through three-dimensional traffic system. In this study, the thrust-to-weight ratio, the wing loading, and the power-to-weight ratio that are major design parameters for the sizing of roadable PAVs were calculated under the constraints of ground roll, climb rate, maximum cruise speed, service ceiling, stall speed. Also, in the sizing process, the study was conducted to determine the design point using the graphs of thrust-to-weight ratio, wing loading, power-to-weight ratio, and brake horse power for the mission profiles considering domestic environments and the FAR PART 23 which is the GA class aircraft certification standard.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.4
• /
• pp.467-473
• /
• 2020

Recently, the number of traffic accidents has also increased with the increase in the penetration rate of cars in Korea. In particular, not only inter-vehicle accidents but also human accidents near crosswalks are increasing, so that more attention to traffic safety around crosswalks are required. In this paper, we propose a system for predicting the safety level around the crosswalk by recognizing an approaching vehicle and estimating the speed of the vehicle using NVIDIA Jetson Nano-class edge devices. To this end, various machine learning models are trained with the information obtained from deep learning-based vehicle detection to predict the degree of risk according to the speed of an approaching vehicle. Finally, based on experiments using actual driving images and web simulation, the performance and the feasibility of the proposed system are validated.