• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1479-1501
• /
• 2020

Dependable functioning of dynamic networks is essential for delivering ubiquitous services. Faults are the root causes of network outages. The comparison diagnosis model, which automates fault's identification, is one of the leading approaches to attain network dependability. Most of the existing research has focused on stationary networks. Nonetheless, the time-free comparison model imposes no time constraints on the system under considerations, and it suits most of the diagnosis requirements of dynamic networks. This paper presents a novel protocol that diagnoses faulty nodes in diagnosable dynamic networks. The proposed protocol comprises two stages, a testing stage, which uses the time-free comparison model to diagnose faulty neighbour nodes, and a disseminating stage, which leverages a Random Linear Network Coding (RLNC) technique to disseminate the partial view of nodes. We analysed and evaluated the performance of the proposed protocol under various scenarios, considering two metrics: communication overhead and diagnosis time. The simulation results revealed that the proposed protocol diagnoses different types of faults in dynamic networks. Compared with most related protocols, our proposed protocol has very low communication overhead and diagnosis time. These results demonstrated that the proposed protocol is energy-efficient, scalable, and robust.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.23 no.4
• /
• pp.347-355
• /
• 2013

A damage estimation method for monopile support structure of offshore wind turbine using modal properties and committee of neural networks is presented for effective structural health monitoring. An analytical model for a monopile support structure is established, and the natural frequencies, mode shapes, and mode shape slopes for the support structure are calculated considering soil condition and added mass. The input to the neural networks consists of the modal properties and the output is composed of the stiffness indices of the support structure. Multiple neural networks are constructed and each individual network is trained independently with different initial synaptic weights. Then, the estimated stiffness indices from different neural networks are averaged. Ten damage cases are estimated using the proposed method, and the identified damage locations and severities agree reasonably well with the exact values. The accuracy of the estimation can be improved by applying the committee of neural networks which is a statistical approach averaging the damage indices in the functional space.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• The Korean Journal of Health Service Management
• /
• v.7 no.3
• /
• pp.177-192
• /
• 2013

The present study investigated the effect of area-specific social networks on urban workers life satisfaction. For this, 356 adults over age 20 were interviewed from June 17th 2013 to June 29th 2013. The findings are as follows: First, the closeness of family network index demonstrates that participants with higher affective support have higher life satisfaction. In addition, stronger extended family network brings more life satisfaction and so does a bigger friendship network. Secondly, the extended family network explains 17.6% of the variance in social networks follows by family network, other network, and friendship network, respectively. The closeness variable of social networks yields statistical significance on all categories of networks. The affective support level in the closeness variable of social networks shows differences as well; family network positively associates with life satisfaction. The purposes of this research are to investigate the actual conditions of urban workers' life satisfaction and the influence of family, extended family, friendship, and other variables on life satisfaction. If social networks have an influence on life satisfaction, to find the domain of social networks that holds the most influences on life satisfaction is an important ground in the process of implementing regional welfare.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.4
• /
• pp.1493-1515
• /
• 2015

A smart meter is one of the core components in Advanced Metering Infrastructure (AMI) that is responsible for providing effective control and monitor of electrical energy consumptions. The multifunction tasks that a smart meter carries out such as facilitating two-way communication between utility providers and consumers, managing metering data, delivering anomalies reports, analyzing fault and power quality, simply show that there are huge amount of data exchange in smart metering networks (SMNs). These data are prone to security threats due to high dependability of SMNs on Internet-based communication, which is highly insecure. Therefore, there is a need to identify all possible security threats over this network and propose suitable countermeasures for securing the communication between smart meters and utility provider office. This paper studies the architecture of the smart grid communication networks, focuses on smart metering networks and discusses how such networks can be vulnerable to security attacks. This paper also presents current mechanisms that have been used to secure the smart metering networks from specific type of attacks in SMNs. Moreover, we highlight several open issues related to the security and privacy of SMNs which we anticipate could serve as baseline for future research directions.

• Journal of Communications and Networks
• /
• v.11 no.6
• /
• pp.626-633
• /
• 2009

Lawful interception (LI) involves legally accessing private communication such as telephone calls or email messages. Numerous countries have been drafting and enacting laws concerning the LI procedures. With the proliferation of portable Internet services such as the IEEE 802.16e wireless mobile networks, surveillance over illegal users is an emerging technical issue in LI. The evermigrating users and their changing IP's make it harder to provide support for seamless LI procedures on 802.16e networks. Few studies, however, on seamless LI support have been conducted on the 802.16e mobile networks environments. Proposed in this paper are a seamless LI architecture and algorithms for the 802.16e networks. The simulation results demonstrate that the proposed architecture improves recall rates in intercepting mobile user, when compared to the existing LI architectures.

• Journal of the Korean Institute of Navigation
• /
• v.21 no.3
• /
• pp.55-66
• /
• 1997

In this paper, applications of neural networks to vibration control of flexible single link robot manipulator are ocnsidered. The architecture of neural networks is a hidden layer, which is comprised of self-recurrent one. Tow neural networks are utilized in a control system ; one as an identifier is called neuro identifier and the othe ra s a controller is called neuro controller. The neural networks can be used to approximate any continuous function to any desired degree of accuracy and the weights are updated by dynamic error-backpropagation algorithm(DEA). To guarantee concegence and to get faster learning, an approach that uses adaptive learning rates is developed by introducing a Lyapunov function. When a flexible manipulator is ratated by a motor through the fixed end, transverse vibration may occur. The motor torque should be controlle dinsuch as way, that the motor is rotated by a specified angle. while simulataneously stabilizing vibration of the flexible manipulators so that it is arrested as soon as possible at the end of rotation. Accurate vibration control of lightweight manipulator during the large body motions, as well as the flexural vibrations. Therefore, dynamic models for a flexible single link manipulator is derived, and LQR controller and nerual networks controller are composed. The effectiveness of the proposed nerual networks control system is confirmed by experiments.

• Journal of Distribution Science
• /
• v.14 no.9
• /
• pp.31-45
• /
• 2016

Purpose - This study develops a simulation model that looks at the dynamics between social capital and business performance in strategic networks to understand their behaviors in relation to each other, and to suggest dynamic relationship strategies. Research design, data, and methodology - Based on existing literature, this study identifies the complex causal loop diagram on social capital and business performance in strategic networks, and converts them into a simulation model for observing how the changes in business environment and relationship dependency affect social capital and business performance. Results - The simulation results showed that, first, the formation in social capital and business performance of networks with low relationship dependency was less affected by the changes in business environment. Second, the formation in social capital and business performance of networks with high relationship dependency was negatively impacted by the changes in business environment. In other words, higher relationship dependency strengthened the impact of changes in business environment on business performance. Conclusions - Thus, this study confirmed that in strategic networks, the changes in business environment and the degree of relationship dependency dynamically affect business performance, and that relationship dependency mediates the degree in which changes in the business environment affect business performance. The results of the simulations were further verified through actual business cases.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.1
• /
• pp.49-59
• /
• 2019

Generative adversarial networks(GANs) have received great attention in the machine learning field for their capacity to model high-dimensional and complex data distribution implicitly and generate new data samples from the model distribution. This paper investigates the model training methodology, architecture, and various applications of generative adversarial networks. Experimental evaluation is also conducted for generating synthetic image dataset for defense using two types of GANs. The first one is for military image generation utilizing the deep convolutional generative adversarial networks(DCGAN). The other is for visible-to-infrared image translation utilizing the cycle-consistent generative adversarial networks(CycleGAN). Each model can yield a great diversity of high-fidelity synthetic images compared to training ones. This result opens up the possibility of using inexpensive synthetic images for training neural networks while avoiding the enormous expense of collecting large amounts of hand-annotated real dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4625-4647
• /
• 2020

The Generative Adversarial Networks, as one of the most creative deep learning models in recent years, has achieved great success in computer vision and natural language processing. It uses the game theory to generate the best sample in generator and discriminator. Recently, many deep learning models have been applied to the security field. Along with the idea of "generative" and "adversarial", researchers are trying to apply Generative Adversarial Networks to the security field. This paper presents the development of Generative Adversarial Networks. We review traditional generation models and typical Generative Adversarial Networks models, analyze the application of their models in natural language processing and computer vision. To emphasize that Generative Adversarial Networks models are feasible to be used in security, we separately review the contributions that their defenses in information security, cyber security and artificial intelligence security. Finally, drawing on the reviewed literature, we provide a broader outlook of this research direction.