• Journal of the Korean Society of Mechanical Technology
• /
• v.20 no.6
• /
• pp.776-782
• /
• 2018

After CNN basic structure was introduced by LeCun in 1989, there has not been a major structure change except for more deep network until recently. The deep network enhances the expression power due to improve the abstraction ability of the network, and can learn complex problems by increasing non linearity. However, the learning of a deep network means that it has vanishing gradient or longer learning time. In this study, we proposes a CNN structure with MLP layer. The proposed CNNs are superior to the general CNN in their classification performance. It is confirmed that classification accuracy is high due to include MLP layer which improves non linearity by experiment. In order to increase the performance without making a deep network, it is confirmed that the performance is improved by increasing the non linearity of the network.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.5
• /
• pp.666-674
• /
• 2013

This paper studied about the design and implementation of tactical communication network simulator in order to obtain tactical communication network parameter, such as link capacity and routing plan, and a number of exceptional cases that may occur during actual deployment by conducting simulation of a large-scale tactical communication networks. This tactical communication network simulator provides equipment models and link models of commercial OPNET simulator for tactical communication network. In addition, 6 types of simulation scenario writings convenience functions and traffic generation models that may occur in situations of tactical communication network environment were implemented in order to enhance user friendliness. By taking advantages of SITL(System-In-The-Loop) function of OPNET, the tactical communication network simulator allows users to perform interoperability test between M&S models and actual equipment in operating simulation of tactical communication network, which is run on software. In order to confirm the functions and performance of the simulator, small-scale of tactical communication network was configured to make sure interoperability between SITL-based equipment and a large-scale tactical communication network was simulated and checked how to cope with traffic generated for each network node. As the results, we were able to confirm that the simulator is operated properly.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.05a
• /
• pp.101-104
• /
• 2001

This paper deals with the speed-up of interval regression neural network. We propose an improved method of adjusting the parameter alpha used in the interval regression neural network to improve the learning speed and regression performance. Finally, we provide numerical examples to evaluate the performance of the proposed method.

• Journal of Korean Library and Information Science Society
• /
• v.44 no.3
• /
• pp.405-428
• /
• 2013

This study tried to explore the relation between research networks(coauthor network, author co-citation network, author bibliographic coupling network) and research performance of Records and Archives Management study in Korea. For the analysis, three basic types of network centrality and three indicators of research performance are used. The summary of this study is as follows: Firstly, there are relations between three centralities and three indicators of research performance in the coauthor network. Secondly, there are relations between betweenness centrality and research performance in the author co-citation/author bibliographic coupling networks. Thirdly, there are relations between three centralities in the each research network. Fourthly, there are not high relations between all centralities of the three research networks.

• Journal of Korea Society of Industrial Information Systems
• /
• v.22 no.5
• /
• pp.17-24
• /
• 2017

Recently Deep-learning, Technologies using Large or Deep Artificial Neural Networks, have Shown Remarkable Performance, and the Increasing Size of the Network Contributes to its Performance Improvement. However, the Increase in the Size of the Neural Network Leads to an Increase in the Calculation Amount, which Causes Problems Such as Circuit Complexity, Price, Heat Generation, and Real-time Restriction. In This Paper, We Propose and Test a Method to Reduce the Number of Network Connections by Effectively Pruning the Redundancy in the Connection and Showing the Difference between the Performance and the Desired Range of the Original Neural Network. In Particular, we Proposed a Simple Method to Improve the Performance by Re-learning and to Guarantee the Desired Performance by Allocating the Error Rate per Layer in Order to Consider the Difference of each Layer. Experiments have been Performed on a Typical Neural Network Structure such as FCN (full connection network) and CNN (convolution neural network) Structure and Confirmed that the Performance Similar to that of the Original Neural Network can be Obtained by Only about 1/10 Connection.

• The Journal of the Korea Contents Association
• /
• v.18 no.2
• /
• pp.188-199
• /
• 2018

The purpose of this study is to grasp the type of network activity and corporate competency based on network theory and to compare and analyze the relationship between these characteristics and business performance. The results showed that the frequency, importance, and reliability of CEO's had a positive (+) effect on business performance, and the mediating effect of corporate competency (industrial information competency, opportunity capture competency, strategic flexibility, and transaction cost reduction) appeared. This study tried to measure business performance by including corporate competency as a direct performance variable of CEO's network activity and found that it is desirable to focus on interacting with the most important network sources and to make efforts to strengthen the qualitative characteristics of network activities rather than expanding the scope of the network in order to improve the performance of CEO's Network Activity.

• Management & Information Systems Review
• /
• v.31 no.2
• /
• pp.61-87
• /
• 2012

In order to improve the value of individuals and the organization's performance, strenuous efforts have been made on improving their human capital and expanding their social capital by members of the organization. Accordingly, the purpose of this study is to verify how members' intimacy network of an organization affects the organization's performance. In this research, they are verified by setting the characteristic of individuals intimacy network as an independent variable, and the job performance and the degrees of cooperation in the network as dependent variables. The results of this study show as follows. First, the size, strength, approach of intimacy network showed significant results. In particular, the first impression when first saw him becomes an important variable. And, that is affected by the approach of intimacy network. Second, the approach of intimacy network has a high impact on the cooperative behavior and job performance in a group which is formed by relationships through face-to-face contract.

• Journal of Digital Contents Society
• /
• v.19 no.4
• /
• pp.801-808
• /
• 2018

The performance test methodology for switch, router or integrated networks that make up the wired network is not clearly defined. Due to smart phones and multimedia traffic, the speed of the network is rapidly increasing to 100Gbps. However, there is no formalized test method, and performance test related parts are still being tested based on old data. Telecommunication companies and companies are testing their requirements in their own BMT, and government or public institutions are requesting performance tests from external organizations. As the demand for the performance test of next generation network equipments and the integrated network increases, analysis of the performance test method should be done as soon as possible. Therefore, in this paper, we analyze the performance test method performed by the institute and propose guidelines that can be applied to all wired network equipment.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.2
• /
• pp.141-156
• /
• 2009

As the TCP is the protocol designed for the wired network that packet loss probability is very low, because TCP transmitter takes it for granted that the packet loss by the wireless network characteristics is occurred by the network congestion and lowers the transmitter's transmission rate, the performance is degraded. The Snoop Protocol was designed for the wired network by putting the Snoop agent module on the BS(Base Station) that connect the wire network to the wireless network to complement the TCP problem. The Snoop agent cash the packets being transferred to the wireless terminal and recover the loss by resending locally for the error occurred in the wireless link. The Snoop agent blocks the unnecessary congestion control by preventing the dupack (duplicate acknowledgement)for the retransmitted packet from sending to the sender and hiding the loss in the wireless link from the sender. We evaluated the performance in the wired/wireless network and in various TCP versions using the TCP designed for the wired network and the Snoop designed for the wireless network and evaluated the performance of the wired/wireless hybrid network in the wireless link environment that the continuous packet loss occur.

• International journal of advanced smart convergence
• /
• v.5 no.1
• /
• pp.8-15
• /
• 2016

Studies have been made for the wireless sensor network protocols by a number of researchers to date. In particular, the studies as to the hierarchical protocol LEACH algorithm was concentrated. Various studies have been derived for the performance of the protocol is based on the LEACH protocol have been made. Improved algorithms have been proposed continuously. On the other hand, The performance comparison and evaluation of the improved algorithm is insufficient. Therefore, we compared the performance for the ML-LEACH (Multi Hop-Layered) and DL-LEACH (Dual Hop-Layered) been derived mainly LEACH. scalability, energy consumption, CH elected, network lifetime were selected as a Performance evaluation items.