• The KIPS Transactions:PartC
• /
• v.17C no.1
• /
• pp.99-108
• /
• 2010

The traffic classification is a preliminary but essentialstep for stable network service provision and efficient network resource management. While various classification methods have been introduced in literature, the payload signature-based classification is accepted to give the highest performance in terms of accuracy, completeness, and practicality. However, the collection and maintenance of up-to-date signatures is very difficult and time consuming process to cope with the dynamics of Internet traffic over time. In this paper, We propose an automatic payload signature generation mechanism which reduces the time for signature generation and increases the granularity of signatures. Furthermore, We describe a signature update system to keep the latest signatures over time. By experiments with our campus network traffic we proved the feasibility of our mechanism.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.9B
• /
• pp.1287-1294
• /
• 2010

The traffic classification is a preliminary and essential step for stable network service provision and efficient network resource management. While a number of classification methods have been introduced in literature, the payload signature-based classification method shows the highest performance in terms of accuracy, completeness, and practicality. However, the payload signature-based method has a significant drawback in high-speed network environment that the processing speed is much slower than other classification method such as header-based and statistical methods. In this paper, We describes various design options to improve the processing speed of traffic classification in design of a payload signature based classification system and describes our selections on the development of our traffic classification system. Also the feasibility of our selection was proved through experimental evaluation on our campus traffic trace.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.3
• /
• pp.149-158
• /
• 2015

The purpose of this paper is to investigate the determinants of new product diffusion. We seek to document and explain systematic features of product diffusion. In this essay, we examine the well-documented empirical regularity that the speed of diffusion has accelerated during the twentieth century. The empirical results show that the main source of acceleration are faster declines in prices. Faster price declines make the product affordable to more consumers within a given period of time. Based on theories of intertemporal price discrimination and learning-by-doing, the association between the speed of adoption and the speed of price decline was explained. Faster price declines are attributed to several product characteristics as well as changes in income distribution. Above all, the introduction of consumer electronic products in more recent years can be regarded as the most important factor in accelerating price declines. Consumer electronic products are technologically different from non-electronic goods, in that semiconductors are important components. As the price of semiconductors has dropped rapidly, the falling production costs can be rapidly incorporated to the price of consumer electronic goods. Furthermore, most of the recently introduced consumer electronic products have network externalities, and many products with network externalities require complementary products. A complementary product becomes more readily or cheaply available as more people have the main product. One major difference between previous studies and this study is that the former focuses only on the factors that operate directly on the speed of adoption, while this study incorporated factors that work through price changes as well as the factors that work directly on the speed of adoption.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.8B
• /
• pp.1170-1178
• /
• 2010

Recently as the number of users and application traffic is increasing on high speed network, the importance of application traffic classification is growing more and more for efficient network resource management. Although a number of methods and algorithms for traffic classification have been introduced, they have some limitations in terms of accuracy and completeness. In this paper we propose an application traffic classification based multi-level architecture which integrates several signature-based methods and behavior algorithm, and analyzes traffic using correlation among traffic flows. By strengthening the strength and making up for the weakness of individual methods we could construct a flexible and robust multi-level classification system. Also, by experiments with our campus network traffic we proved the performance and validity of the proposed mechanism.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.8
• /
• pp.335-342
• /
• 2013

Nowadays, the traffic type and behavior are extremely diverse due to the growth of network speed and the appearance of various services on Internet. For efficient network operation and management, the importance of application-level traffic identification is more and more increasing in the area of traffic analysis. In recent years traffic identification methodology using statistical features of traffic flow has been broadly studied. However, there are several problems to be considered in the identification methodology base on statistical features of flow to improve the analysis accuracy. In this paper, we recognize these problems by analyzing the ground-truth traffic and propose the solution of these problems. The four problems considered in this paper are the distance measurement of features, the selection of the representative value of features, the abnormal behavior of TCP sessions, and the weight assignment to the feature. The proposed solutions were verified by showing the performance improvement through experiments in campus network.

• Journal of Internet Computing and Services
• /
• v.10 no.2
• /
• pp.65-73
• /
• 2009

IEEE802.11 Wireless LAN(WLAN) is being widely used in public space such as airport, and increases the networking boundary in campus and enterprise, and it has lastly attracted considerable attention for mesh network and converged network with other 3G mobile communication networks. In WLAN, load balancing among Access Points(AP) is an important issue for efficient resource management or supporting the Quality of Service(QoS) of traffic, but most researches focused on the AP selection in network entry or roaming of Stations(STA). In this paper, we propose an AP-Initiated Flow Redirection(FR) for AP load balancing by monitoring AP's availability in the true sense. When the AP's resource becomes almost saturated, that is used more than a specific threshold, the AP queries the roaming possible neighbor APs about their availability and calculates the distribution of traffic load with statistical methods such as entropy or chi-square. Finally, the AP decides flows and new APs for redirection and performs it. Our simulation results show that our FR mechanism increases the performance in the various views.

• KIISE Transactions on Computing Practices
• /
• v.22 no.8
• /
• pp.375-380
• /
• 2016

The proliferation of online social networking services (OSNSs) and smartphones has enabled people to easily make friends with a large number of users in the online communities, and interact with each other. This leads to an increase in the usage rate of OSNSs. However, individuals who have immersed into their digital lives, prioritizing the virtual world against the real one, become more and more isolated in the physical world. Thus, their socialization processes that are undertaken only through lots of face-to-face interactions and trial-and-errors are apt to be neglected via 'Add Friend' kind of functions in OSNSs. In this paper, we present a friend recommendation system based on the on/off-line contextual information for the OSNS users to have more serendipitous offline interactions. In order to accomplish this, we modeled both offline information (i.e., place visit history) collected from a user's smartphone on a 3D tensor, and online social data (i.e., friend relationships) from Facebook on a matrix. We then recommended like-minded people and encouraged their offline interactions. We evaluated the users' satisfaction based on a real-world dataset collected from 43 users (12 on-campus users and 31 users randomly selected from Facebook friends of on-campus users).

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.11
• /
• pp.105-114
• /
• 2020

In this paper, we address security issues in 5G non-public networks for industrial applications. In contrast to public networks that offer mobile network services to the general public, 5G non-public networks provide 5G network services to a clearly defined user organization or groups of organizations, and they are deployed on the organization's defined premises, such as a campus or a factory. The main goal of this paper is to derive security threats and potential security requirements in the case that 5G non-public networks are built for discrete and process industries according to the four deployment models of 5G-ACIA (5G Alliance for Connected Industries and Automation). In order to clarify the scope of this paper, we express the security toolbox to be applied to 5G non-public networks in the form of the defense in depth concept. Security issues related to general 5G mobile communication services are not within the scope of this paper. We then derive the security issues to consider when applying the 5G-ACIA deployment models to the industrial domain. The security issues are divided into three categories, and they are described in the order of overview, security threats, and potential security requirements.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.5
• /
• pp.1-7
• /
• 2012

In this paper, we design and implement the simulator for WAN storage replication model performance evaluation in cloud computing environment. Each cloud of simulator is composed of virtual machine emulator and storage emulator. The virtual machine emulator is composed of read/write ratio module, the read/write sequence combination module, and the read/write request module. The storage emulator is composed of storage management module, data transfer module, read/write operations module, and overhead processing module. Using the simulator, we evaluate performance of migration scheme, pre-copy and the post-copy, considering about read/write ratio, network delay, and network bandwidth. Through simulation, we have confirmed that the average migration time of pre-copy was decreased proportional to the read operation. However, average migration time of post-copy was on the increase. Also, the average migration time of post-copy was increased proportional to the network delay. However, average migration time of pre-copy was shown uniformly. Therefore, we show that pre-copy model more effective to reduce the average migration time than the post-copy model. The average migration time of pre-copy and post-copy were not affected by the change of network bandwidth. Therefore, these results show that selects the storage replication model to be, the network bandwidth know not being the important element.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.57-65
• /
• 2018

Conventional RF signal-based indoor localization techniques such as BLE or Wi-Fi based fingerprinting method show considerable localization errors even in small-scale indoor environments due to unstable received signal strength(RSS) of RF signals. Therefore, it is difficult to apply the existing RF-based fingerprinting techniques to large-scale indoor environments such as airports and department stores. In this paper, instead of RF signal we use the geomagnetic sensor signal for indoor localization, whose signal strength is more stable than RF RSS. Although similar geomagnetic field values exist in indoor space, an object movement would experience a unique sequence of the geomagnetic field signals as the movement continues. We use a deep neural network model called the recurrent neural network (RNN), which is effective in recognizing time-varying sequences of sensor data, to track the user's location and movement path. To evaluate the performance of the proposed geomagnetic field based indoor positioning system (IPS), we constructed a magnetic field map for a campus testbed of about $94m{\times}26$ dimension and trained RNN using various potential movement paths and their location data extracted from the magnetic field map. By adjusting various hyperparameters, we could achieve an average localization error of 1.20 meters in the testbed.