• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2292-2294
• /
• 2002

This paper presents the network of Distributed Control System(DCS) considering specification of nuclear power plant. The network is composed of field network, control network and information network. The protocol of control network is ring type and it is compared to ethernet type. This paper proposes the structure of DCS, the protocol of each network and analyzes the network traffic along data capacity of field network, control network, information network and the network performance. Network II.5 is used as traffic simulation tool.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.11
• /
• pp.3584-3602
• /
• 2022

NFV realizes flexible and rapid software deployment and management of network functions in the cloud network, and provides network services in the form of chained virtual network functions (VNFs). However, using VNFs to provide quality guaranteed services is still a challenge because of the inherent difficulty in intelligently scaling VNFs to handle traffic fluctuations. Most existing works scale VNFs with fixed-capacity instances, that is they take instances of the same size and determine a suitable deployment location without considering the cloud network resource distribution. This paper proposes a traffic forecasted assisted proactive VNF scaling approach, and it adopts the instance capacity adaptive to the node resource. We first model the VNF scaling as integer quadratic programming and then propose a proactive adaptive VNF scaling (PAVS) approach. The approach employs an efficient traffic forecasting method based on LSTM to predict the upcoming traffic demands. With the obtained traffic demands, we design a resource-aware new VNF instance deployment algorithm to scale out under-provisioning VNFs and a redundant VNF instance management mechanism to scale in over-provisioning VNFs. Trace-driven simulation demonstrates that our proposed approach can respond to traffic fluctuation in advance and reduce the total cost significantly.

• Journal of information and communication convergence engineering
• /
• v.9 no.3
• /
• pp.325-330
• /
• 2011

Due to the wide growth and popularity of social network website, large numbers of users discover these social network sites are a place where they can be able to spend their leisure time sharing interests, sharing ideas freely, sharing personal experience, and also to search for new friends or partners. These websites give an opportunity for its users to socialize with new people and to keep in touch or reconnect with current or old friends and families across disperse continents, which traditionally replace the common traditional methods. These social network websites need accurate and careful investigations and findings on the usability issues for effective interactivity and more usability. However, little research might have previously invested on the usability of these social network websites. Therefore, we propose a new framework to study and test the usability of these social network sites. We namely call our framework "Interactivity". This framework will enable developers to assess the usability of the social network sites. It will provide an overview of the user's behavior while interacting in these social network websites. Performance of the framework will be performed using Camtasia software. This software will entirely capture the interactivity of users including the screen and the movements, which the screen and the motion of the user action will undergo to analysis at the end of our research.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.8B
• /
• pp.1421-1425
• /
• 1999

A restoration scheme is presented when a failure occurs at nodes or links on the ATM network. Each node on the network contains the same topology information for resources on the network by Exchanging NSE(Network State Element) which has the network state information through RCC(Routing Control Channel). We also propose a new establishing algorithm that set up a new alternate path reflecting the state of the network changed dynamically at the source node. In the result of applying our scheme, the rates of restoration can be improved and the number of messages decreased so that fast recovering could be possible by reducing traffics on the network by lowering the rates of the call-setup failure at the stage of CAC(Connection Admission Control).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.1B
• /
• pp.43-50
• /
• 2011

Recently, Hacking Attacks are appearing as a various Attack techniques with evolution of the Network and most of the network through a various Security Systems are responding to an attack. In addition, it should be placed adding the Security Systems to protect the Internal Network's Information Assets from External attacks. But, The use of Security Systems and Network deployment inside the network makes a significant impact on Security and Performance. Therefore, In this paper, it will be to analyze the Performance Degradation Factors of the Internal Network according to the Security System's use and placement. In a future, This paper is expected to serve as a valuable information for the Network Performance and Security improvements.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.2
• /
• pp.159-166
• /
• 2018

The proliferation of information technologies made it possible to produce information products of different versions at much lower cost comparing to traditional physical products. Thus it is common for information product manufacturers to consider vertically differentiated product line for more profit through improved market coverage. Another salient characteristic of most information product is network externality. Existing researches dealing with vertical differentiation and network externality usually assumed oligopolistic market where vertically differentiated products are provided by competing companies, respectively. Moreover, they analyzed the essentially dynamic characteristic of network externality statically. In this study, different from the previous researches, the vertical differentiation strategy of a monopolistic company under network externality is dynamically analyzed. We used a two-period model to accommodate the dynamic feature of network externality. Based on the two-period model, the profit maximizing solutions are analyzed. The results showed that a monopolistic company has no incentive to differentiate products vertically when the network externality is absent. On the contrary, when the network externality exists, the monopolistic company can derive more profit by vertically differentiating the product line. It is also shown that, for more profit, the monopolistic company should keep the quality difference between the high quality product and the low quality product as greater as possible.

• KSCI Review
• /
• v.1 no.2
• /
• pp.73-94
• /
• 1995

In this paper, design of efficient NDB(Network Data Base) for the activation plan of chain information network. The DB structure build up, logical structure, store structure, physical structure, the data express for one's record, and the express using linked in the releation of data. Also express as hierarchical model on the DSD(Data Structure Diagram) from the database with logical structure. Each node has express on record type, the linked in course of connective this type, the infuence have efficent of access or search of data, in the design for connection mutually a device of physical, design for database, and construction a form of store for logical. Also activation of chain information network of efficent, using POS(Point Of Sale) system in OSI(Open Systems Interconnection) environment for network standardization, and build up network a design for system.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.225-230
• /
• 2022

The intelligent home network system integrates various devices in the home into one communication network to provide information sharing, control, and operation environment between devices. This intelligent home network system operates around a home server. Home appliances in the era of the 4th industrial revolution will have numerous home servers in logical areas as the intelligent home network in the home accelerates. Therefore, the need for systematic management of home servers is emerging. We propose an agent system for efficient intelligent smart home server management. The agent system monitors the home server and operating environment for home server management of the intelligent smart home network. By referring to this monitored information, the service module of the home server is managed, and the home server is dealt with whether it is normal or not. In addition, by referring to the information collected by the service agent created in the group management server while migrating the home server, it is possible to deal with integrated meter reading, crime prevention, and topics. And when a new service is applied to the home server, it is registered in the management server and distributed to the home server through the agent, so that the intelligent smart home network can be efficiently managed.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1051-1055
• /
• 2003

Policies are collections of general principles specifying the desired behavior and state of a system. Network management is mainly carried out by following policies about the behavior of the resources in the network. Policy-based (PB) network management supports to manage distributed system in a flexible and dynamic way. This paper focuses on configuration management based on Internet Engineering Task Force (IETF) standards. Network security approaches include the usage of intrusion detection system to detect the intrusion, building firewall to protect the internal systems and network. This paper presents how the policy-based framework is collaborated among the network security systems (intrusion detection system, firewall) and intrusion detection systems are cooperated to detect the intrusions.