Youngjoon Won

Fine‐grained traffic classification based on functional separation

SUMMARY Current efforts to classify Internet traffic highlight accuracy. Previous studies have focused on the detection of major applications such as P2P and streaming applications. However, these applications can generate various types of traffic which are often considered as minor and ignorant traffic portions. As network applications become more complex, the price paid for not concentrating on minor traffic classes is in reduction of accuracy and completeness. In this context, we propose a fine-grained traffic classification scheme and its detailed method, called functional separation. Our proposal can detect, according to functionalities, different types of traffic generated by a single application and should increase completeness by reducing the amount of undetected traffic. We verify our method with real-world traffic. Our performance comparison against existing DPI-based classification frameworks shows that the fine-grained classification scheme achieves consistently higher accuracyand completeness. Copyright

Disasters seen through Flickr cameras

Collecting aftermath information after a wide-area disaster is a crucial task in the disaster response that requires important human resources. We propose to assist reconnaissance teams by extracting useful data sent by the users of social networks that experienced the disaster. In particular we consider the photo sharing website Flickr as a source of information that allows one to evaluate the disaster aftermath. We propose a methodology to detect major event occurrences from the behavior of Flickr users and describe the nature of these events from the tags they post on the Flickr website. Our experiments using two study cases, namely, the Tohoku earthquake and tsunami and the Tuscaloosa tornado, reveals the value of the data published by Flickr users and highlight the value of social networks in disaster response.

Efficient recommendation methods using category experts for a large dataset

Abstract Neighborhood-based methods have been proposed to satisfy both the performance and accuracy in recommendation systems. It is difficult, however, to satisfy them together because there is a tradeoff between them especially in a big data environment. In this paper, we present a novel method, called a CE method, using the notion of category experts in order to leverage the tradeoff between performance and accuracy. The CE method selects a few users as experts in each category and uses their ratings rather than ordinary neighbors’. In addition, we suggest CES and CEP methods, variants of the CE method, to achieve higher accuracy. The CES method considers the similarity between the active user and category expert in ratings prediction, and the CEP method utilizes the active user’s preference (interest) on each category. Finally, we combine all the approaches to create a CESP method, considering similarity and preference simultaneously. Using real-world datasets from MovieLens and Ciao, we show that our proposal successfully leverages the tradeoff between the performance and accuracy and outperforms existing neighborhood-based recommendation methods in coverage. More specifically, the CESP method provides 5% improved accuracy compared to the item-based method while performing 9 times faster than the user-based method.

An approach for failure recognition in IP-based industrial control networks and systems

Industrial control networks (ICNs) and systems support robust communications of devices in process control or manufacturing environments. ICN proprietary protocols are being migrated to Ethernet/IP networks in order to merge various different types of networks into a single common network. ICNs are deployed in mission-critical operations, which require a maximum level of network stability. Network stability is often described using several categories of network performance quality-of-service metrics, such as throughput, delay, and loss measurements. The question arises as to whether these network performance metrics are sufficient to run valuable diagnostics of ICN components and their communications. Any abnormal decision with respect to typical IP traffic behavior does not necessarily coincide with ICN fault cases. A precise and specific diagnostic technique for ICNs is required to remove the uncertainty in detecting problems. However, existing Ethernet/IP diagnosis tools have not been able to fully handle fault symptoms and mainly focus on network diagnostics rather than process or device diagnostics. This paper demonstrates that the absence of advanced fault diagnosis techniques leads to the development of new methodologies that are suitazble for ICN. We describe unique traffic characteristics and categorize the faults of ICN. We also propose a fault diagnosis, prediction, and adaptive decision methodologies, and verify them with real-world ICN data from the steel-making company POSCO. Our experience in developing the fault diagnosis system provides a firm guideline to understand the fault management mechanisms in large ICNs. Copyright

Measurement and analysis of application-level crowd-sourced LTE and LTE-A networks

LTE-Advanced (LTE-A) theoretically can provide better network performance than 4G LTE. Mobile network operators around the world are eager to deploy LTE-A to attract more subscribers. However, is it really making difference to the user experience compared to the existing LTE service? To investigate this question, we collected the cellular network performance log from 111 user smartphones on 3G, LTE, and LTE-A. For in-depth analysis, we also asked for privacy information, such as data plan, subscribed network, monthly payment details, and etc. In addition, we also collected the WiFi network performance log from 83 user smartphones to compare with the cellular ones. By analyzing the collected log, we observed that (i) LTE-A was faster than LTE only for download bandwidth by 14.1%, yet the users pay on average 13.6% more for LTE-A; (ii) the mVoIP traffic was blocked by all carriers when the users exceed their mVoIP quota; (iii) the subscribed data plan showed no discrimination in network performance.

Measuring auto switch between Wi-Fi and mobile data networks in an urban area

To preserve consistent throughput, smartphones are equipped with a network switch feature (handover in heterogeneous networks). Frequent switching is often blamed to be a QoE downgrader in populated areas. In this paper, we measured auto switch occurrences between Wi-Fi and mobile data networks. We deployed an Android monitoring application for 89 participants and collected network status information up to 10 days long. We observed that auto switch occurred on average 2.53 times per hour and RTT decreased as the smartphone preferred to stay in Wi-Fi. Also, 68% of all users connected to Wi-Fi longer than the mobile data networks.

Analyzing network privacy preserving methods: a perspective of social network characteristics

This paper investigates structural and property changes via several privacy preserving methods (anonymization) for social network. We observe inconsistency of privacy preserving methods in social network analysis.

Is LTE-Advanced really advanced?

LTE-Advanced (LTE-A) theoretically can provide better network performance than 4G LTE. Mobile carriers around the globe are eager to deploy LTE-A to attract more subscribers. However, is it really making difference to the user experience compared to the existing LTE service? To investigate this question, we collected the network performance log from 111 user smartphones on 3G, LTE, and LTE-A. For in-depth analysis, we also asked for privacy information, such as data plan, subscribed network, monthly payment details, and etc. By analyzing the collected log, we observed that (i) LTE-A was faster than LTE only for download bandwidth by 14.1%, yet the users pay on average 13.6% more for LTE-A; (ii) the mVoIP traffic was blocked by all carriers when the users exceed their mVoIP quota; (iii) the subscribed data plan showed no discrimination in network performance.

Mass-Count Disparity in Mobile Traffic

Mass-count disparity is a basis of the elephants and mice phenomenon in Internet traffic analysis. Mobile applications tend to minimize transmission overhead by reducing object size. Assuming the mice get smaller, we first look at properties of mass-count disparity in the smart device traffic. We find the existence of elephants and represent an accurate inequality measure using the Gini coefficient where the heavy-tail property is not clearly visible in the trace. The Gini coefficients range from 0.71 (web) to 0.98 (application market traffic), implying a heavy inequality distribution toward 1. The cutoff point of elephants is 1.55 MB that is even comparable with small size photos. Our hypothesis from the early analysis indicates that every mobile user is potentially generating elephant flows. We observe that a significant stance of application market traffic is responsible for such phenomenon.