Songkuk Kim

On the Impact of Layer-Splitting for Cloud-Based SVC Streaming

Video streaming has dominated traffic usage over the Internet. With the development of heterogeneous mobile devices, various streaming environments with different display sizes and network bandwidths have accelerated video streaming to the masses. To accommodate the increasing demand for video, streaming providers have made efforts to build services based on scalable video coding (SVC) over the cloud infrastructure, which leverages elastic amounts of resources and network bandwidth on the fly. In this paper, we suggest a split-layer SVC-based streaming architecture optimized for a cloud environment. By separately controlling streaming instances with awareness of clients device ratio amongst smartphone, tablet, and PC, we improve resource utilization and save network bandwidth. We also evaluate how our proposed system can improve the total cost for providing a video streaming service on a cloud infrastructure. We carry out extensive experiments to evaluate the performance in multiple real-life scenarios by simulating streaming services on (1) a mobile-client-heavy channel, (2) a PC-client-heavy channel, and (3) a device-balanced channel.

Virtualized ANR to Manage Resources for Optimization of Neighbour Cell Lists in 5G Mobile Wireless Networks

In future, more devices such as wearable devices will be connected to the networks. This will increase simultaneous handovers. The coverage of a cell will be small because a superhigh frequency used in 5G wireless networks does not propagate very far. This trend will increase the number of neighbour cell lists and it will accelerate the change of neighbour cell lists since the coverage of cells can be altered by the environment. Meanwhile, the ANR technology will be essential in 5G networks. Since the network environment in the future is not similar to the present, the strategy of ANR should also be different from the present. First, since practical neighbour cell lists in each cell are changed frequently and individually, it is necessary to optimize them frequently and individually. Second, since the neighbour cell lists in each cell are not changed similarly, it is necessary to operate ANR flexibly. To respond to these issues, we propose to use network function virtualization (NFV) for ANR. To evaluate the proposed strategies, we measured additional resource consumption and the latency of handover if neighbour cell lists are not optimized when UEs perform handover simultaneously. These experiments are conducted using Amarisoft LTE-100 Platform.

Optimal Multi-Interface Selection for Mobile Video Streaming in Efficient Battery Consumption and Data Usage

With the proliferation of high-performance, large-screen mobile devices, users’ expectations of having access to high-resolution video content in smooth network environments are steadily growing. To guarantee such stable streaming, a high cellular network bandwidth is required; yet network providers often charge high prices for even limited data plans. Moreover, the costs of smoothly streaming high-resolution videos are not merely monetary; the device’s battery life must also be accounted for. To resolve these problems, we design an optimal multi-interface selection system for streaming video over HTTP/TCP. An optimization problem including battery life and LTE data constraints is derived and then solved using binary integer programming. Additionally, the system is designed with an adoption of split-layer scalable video coding, which provides direct adaptations of video quality and prevents out-of-order packet delivery problems. The proposed system is evaluated using a prototype application in a real, iOS-based device as well as through experiments conducted in heterogeneous mobile scenarios. Results show that the system not only guarantees the highest-possible video quality, but also prevents reckless consumption of LTE data and battery life.

25-Hydroxyvitamin D status and bone mineral density of korean women

25-HYDROXYVITAMIN D STATUS AND BONE MINERAL DENSITY OF KOREAN WOMEN. JH KIM~ SW K1M, SJ MOON, SK LIM. Dept. of Food and Nutrition, Dept. of Internal Medicine, Yonsei University. Seoul, Korea. Level of serum 25-hydroxyvitamin D(25-OHD) in 179 Korean women aged from 20 to 75 years old was investigated by HPLC (High Pressure Liquid Chromatography). Related biochemical indices such as iPTH, alkaline phosphatase, creatinine, albumin, Ca, Mg,I P were also measured. Bone mineral density (BMD) of the nondominant forearm was measured by single-photon absorptiometry in total subjects at a distal radius site. This study attempted to define reference data for the distribution of vitamin D status and to explore the relationship between vitamin D status and BMD in Korean women so that it may suggest the important role of vitamin D in Korean women. The mean serum 25-OI-[I) level was 25.Sng/ml. Serum 25-OHD was inversely related to iPTH and alkaline phosphatase. There were significant changes in serum 25-OHD level from the preto the post-menopausal women. Significant seasonal variation of serum 25-OHD and iPTH were noted in 26 selected subjects. Among the total subjects, 16.5% showed vitamin D defieieney(i.e, s-25-OHD <10ng/ml). Mean of BMD levels at distalradius decreased significantly with aging. In linear regression models, the slope of the bone loss was significantly greater in women aged 50-59 years compared with women aged over 60 years old. There were significant correlation between BMD and serum 2,5-OHD (p<0.001). Mean of age-adjusted BMD(Z-score) levels were lower in women who had past or current low Ca and vitamin D intake. To define the potential role of subclinical vitamin D deficiency I in postmenopausal bone loss, we analyzed the levels of circulating serum 25-OHD in 33 post-menopausal women screened for osteoporosis and age-matched controls. In most cases they had significantly lower levels of serum 25-OHD, and increased iPTH and alkaline phosphatase. Low serum 25-OI-ID level(<10ng/ml) was founded in 26 osteoporotic subjects(74.3%). Logistic analysis revealed that serum 25-OHD was the major determinant of the observed postmenopausal osteoporosis.

Measuring the eccentricity of items

The long-tail phenomenon tells us that there are many items in the tail. However, not all tail items are the same. Each item acquires different kinds of users. Some items are loved by the general public, while some items are consumed by eccentric fans. In this paper, we propose a novel metric, item eccentricity, to incorporate this difference between consumers of the items. Eccentric items are defined as items that are consumed by eccentric users. We used this metric to analyze two real-world datasets of music and movies and observed the characteristics of items in terms of eccentricity. The results showed that our defined eccentricity of an item does not change much over time, and classified eccentric and noneccentric items present significantly distinct characteristics. The proposed metric effectively separates the eccentric and noneccentric items mixed in the tail, which could not be done with the previous measures, which only consider the popularity of items.

Design and Analysis of Virtualized Caching Service on Cellular Infrastructure

With the large increases in video content, mobile network operators (MNOs) are persistently seeking to lower their content access latency and backhaul traffic. To mitigate these challenges, operators have started developing an operator-owned content caching system. However, in such networks, it is difficult to provide cost-effective and reliable cache performance for dynamically changing traffic using traditional proprietary network equipment. Meanwhile, network function virtualization (NFV) technology has become a key enabler in providing new business models with reduced costs and enhanced service agility to mobile networks. In addition, dynamic scaling of virtualized network functions can be used to leverage adaptable management of network services on the fly. In this study, we introduce a system architecture of MNO-driven virtualized caching-service based on an NFV standard platform that uses an integer linear programming (ILP)-based centralized optimization algorithm. Our proposed system employs mainly two techniques: a hierarchical caching model and a hybrid (centralized and distributed) orchestration scheme to optimize cache function with fine-grained control. Through experiments, we show that the conceptual MNO caching service is an effective strategy for reducing the traffic burden and enhancing the quality of service (QoS) for streaming video.

The virtualized MPTCP proxy performance in cellular network

For massive traffic handling in the cellular network, network function virtualization (NFV) is considered to be the most cost-efficient solution in the 5G networks. Since NFV decouples the network function from the underlying hardware, the purpose-built machines can be replaced by the commodity hardware. However, NFV might suffer from the very fact that it is a solely software-based solution. The objective of this paper is to find out the NFV performance issue in cellular network. Also, we want to investigate whether NFV is comparable with MPTCP connections. Since not many servers are MPTCP-enabled, a SOCKS proxy is usually deployed in between to enable MPTCP connections. We regarded a virtualized proxy as an NFV instance and set up two types of virtualized SOCKS proxies, one as KVM and the other as docker. We also tried to find out if there is a performance difference between hypervisor-based and container-based virtualization in our setting. As the results show, the docker proxy performs better than the KVM proxy. In terms of resource consumption, for example, the docker utilized 31.9% of host CPU, whereas the KVM consumed 36.9% when both of them handling 2,000 concurrent requests. The throughput comparison of different TCP connections reflects the characteristics of MPTCP flow that performs best in a long and large flow. The latency between the server and the proxy determined the throughput of MPTCP with a virtualized proxy. If the latency between the server and the proxy gets larger (RTT 100ms), the MPTCP proxy throughput of all three different flow got worse than the single TCP connections, whether it is a short flow (1KB) or a long flow (164MB). However, if the latency is in the middle range (RTT 50ms), the MPTCP proxy throughput of a short (1KB) and medium (900KB) flow works poorly, but a long flow (164MB) still works better than the single TCP connections.

Multivariate Multiple Regression Models for a Big Data-Empowered SON Framework in Mobile Wireless Networks

In the 5G era, the operational cost of mobile wireless networks will significantly increase. Further, massive network capacity and zero latency will be needed because everything will be connected to mobile networks. Thus, self-organizing networks (SON) are needed, which expedite automatic operation of mobile wireless networks, but have challenges to satisfy the 5G requirements. Therefore, researchers have proposed a framework to empower SON using big data. The recent framework of a big data-empowered SON analyzes the relationship between key performance indicators (KPIs) and related network parameters (NPs) using machine-learning tools, and it develops regression models using a Gaussian process with those parameters. The problem, however, is that the methods of finding the NPs related to the KPIs differ individually. Moreover, the Gaussian process regression model cannot determine the relationship between a KPI and its various related NPs. In this paper, to solve these problems, we proposed multivariate multiple regression models to determine the relationship between various KPIs and NPs. If we assume one KPI and multiple NPs as one set, the proposed models help us process multiple sets at one time. Also, we can find out whether some KPIs are conflicting or not. We implement the proposed models using MapReduce.

Analysis of Key Factors for High Yield AMOLED Display

To figure out the key factors of the reliability of AMOLED displays, we analyzed the luminance degradation based on a large-scale simulation. The initial non-uniformity and aging parameters were incorporated into the luminance decay model. For each set of parameters, we simulated the luminance degradation of 10,000 panels for 100,000 operating hours and analyzed by two criteria; contrast deviation and luminance variation of panels to determine failures. Stability of TFTs is the most significant parameter for realizing highly reliable AMOLED display. Stability of OLEDs is also potential factor when which of TFTs is sufficient.