Skip to main content
main-content

Über dieses Buch

This book presents the combined proceedings of the 12th KIPS International Conference on Ubiquitous Information Technologies and Applications (CUTE 2017) and the 9th International Conference on Computer Science and its Applications (CSA2017), both held in Taichung, Taiwan, December 18 - 20, 2017.

The aim of these two meetings was to promote discussion and interaction among academics, researchers and professionals in the field of ubiquitous computing technologies.

These proceedings reflect the state of the art in the development of computational methods, involving theory, algorithms, numerical simulation, error and uncertainty analysis and novel applications of new processing techniques in engineering, science, and other disciplines related to ubiquitous computing.

James J. (Jong Hyuk) Park received Ph.D. degrees in Graduate School of Information Security from Korea University, Korea and Graduate School of Human Sciences from Waseda University, Japan. From December, 2002 to July, 2007, Dr. Park had been a research scientist of R&D Institute, Hanwha S&C Co., Ltd., Korea. From September, 2007 to August, 2009, He had been a professor at the Department of Computer Science and Engineering, Kyungnam University, Korea. He is now a professor at the Department of Computer Science and Engineering and Department of Interdisciplinary Bio IT Materials, Seoul National University of Science and Technology (SeoulTech), Korea. Dr. Park has published about 200 research papers in international journals and conferences. He has been serving as chair, program committee, or organizing committee chair for many international conferences and workshops. He is a steering chair of international conferences – MUE, FutureTech, CSA, CUTE, UCAWSN, World IT Congress-Jeju. He is editor-in-chief of Human-centric Computing and Information Sciences (HCIS) by Springer, The Journal of Information Processing Systems (JIPS) by KIPS, and Journal of Convergence (JoC) by KIPS CSWRG. He is Associate Editor / Editor of 14 international journals including JoS, JNCA, SCN, CJ, and so on. In addition, he has been serving as a Guest Editor for international journals by some publishers: Springer, Elsevier, John Wiley, Oxford Univ. press, Emerald, Inderscience, MDPI. He got the best paper awards from ISA-08 and ITCS-11 conferences and the outstanding leadership awards from IEEE HPCC-09, ICA3PP-10, IEE ISPA-11, PDCAT-11, IEEE AINA-15. Furthermore, he got the outstanding research awards from the SeoulTech, 2014. His research interests include IoT, Human-centric Ubiquitous Computing, Information Security, Digital Forensics, Vehicular Cloud Computing, Multimedia Computing, etc. He is a member of the IEEE, IEEE Computer Society, KIPS, and KMMS.

Vincenzo Loia (BS ‘85, MS ‘87, PhD ‘89) is Full Professor of Computer Science. His research interests include Intelligent Agents, Ambient intelligence, Computational Intelligence. Currently he is Founder & Editor-in-chief of “Ambient Intelligence and Humanized Computing”, and Co-Editor-in-Chief of “Softcomputing”, Springer-Verlag. He is Chair of the Task Forces “Intelligent Agents” and “Ambient Intelligence” IEEE CIS ETTC. He has been Chair the Emergent Technical Committe "Emergent Technology", IEEE CIS Society and Vice-Chair of Intelligent Systems Applications Technical Committee. He has been author of more than 200 scientific works, Editor/co-editor of 4 Books, 64 journal papers, 25 book chapters, and 100 conference papers. He is Senior member of the IEEE, Associate Editor of IEEE Transactions on Industrial Informatics, and Associate Editor of IEEE Transactions on Systems, Man, and Cybernetics: Systems. Many times reviewers for national and international projects, Dr. Loia is active in the research domain of agents, ambient intelligence, computational intelligence, smartgrids, distributed platform for enrich added value.

Gangman Yi in Computer Sciences at Texas A&M University, USA in 2007, and doctorate in Computer Sciences at Texas A&M University, USA in 2011. In May 2011, he joined System S/W group in Samsung Electronics, Suwon, Korea. He joined the Department of Computer Science & Engineering, Gangneung-Wonju National University, Korea, since March 2012. Dr. Yi has been researched in an interdisciplinary field of researches. His research focuses especially on the development of computational methods to improve understanding of biological systems and its big data. Dr. Yi actively serves as a managing editor and reviewer for international journals, and chair of international conferences and workshops.

Yunsick Sung received his B.S. degree in division of electrical and computer engineering from Pusan National University, Busan, Korea, in 2004, his M.S. degree in computer engineering from Dongguk University, Seoul, Korea, in 2006, and his Ph.D. degree in game engineering from Dongguk University, Seoul, Korea, in 2012. He was employed as a member of the researcher at Samsung Electronics between 2006 and 2009. He was the plural professor at Shinheung College in 2009 and at Dongguk University in 2010. His main research interests are many topics in brain-computer Interface, programming by demonstration, ubiquitous computing and reinforcement learning. His Journal Service Experiences is Associate Editor at Human-centric Computing and Information Sciences, Springer (2015- Current).

Inhaltsverzeichnis

Frontmatter

An Examination of the Effect of the Contextual UI Design Quality of Mobile Shopping Applications on the Loyalty of Users to the Applications

Because the typical display space of today’s mobile devices is so small, most mobile applications have not provided users with enough information or UI (user interface) design components, ever since mobile devices were launched back in the late 1990s. According to a related literature review, the UI design of applications, done from a design perspective to provide users with enough UI design components to deliver data, is of a high contextual UI design quality. Yet, there is very little research in the information systems (IS) field on the effects of contextual UI design quality on the ease of use of mobile shopping applications, or on the loyalty of users to the applications. The main research goals of this study are as follows: (1) to examine the direct effects of the contextual UI design quality of mobile shopping applications on the ease of use of the applications and on the loyalty of users to the applications and (2) to find the relationship between the ease of use of shopping applications and the loyalty of users to the applications. After conducting a survey, this study analyzed the data with Structural Equation Modeling (SEM). The results showed that the contextual UI design quality of mobile shopping applications have significant direct effects on the ease of use of the applications and on the loyalty of users to the applications. In addition, the relationship between the ease of use of such applications and the loyalty of users to the applications was identified.

Wonjin Jung

Securing Intelligent Vehicular Ad Hoc Networks: A Survey

Vehicular ad hoc networks (VANETs) is an intelligent transportation system that provides wireless communication between vehicles and different objects in the road to increase efficiency and human safety using various applications. However, all this attractive features of VANETs will increase security risks and privacy problems if security attacks is not studied and analyzed thoroughly and completely. This paper discuss VANETs’ features, and structures. In addition, it list different security attacks and provide a unique classification for them. Finally, it goes through security architectures and schemes used in VANETs.

Wedad Ahmed, Mourad Elhadef

An Approach of Test Case Generation for Spreadsheet Cells

Spreadsheet is one of the most commonly programming environment, widely adopted for data processing, template building and decision making. However, using spreadsheet often leads to serious consequences because of all kinds of errors in it. This paper presents a tool for test case generation based on data mutation and code smells. Including two steps, detecting sheets types and spreadsheet cells test cases generation. And we perform an experiment with our method, the result demonstrates the effectiveness of our test case generation method. At last, the future work of spreadsheet debugging technology is proposed.

Bo Yang, Qian Yu

Bearing Fault Diagnosis Based on Convolutional Neural Networks with Kurtogram Representation of Acoustic Emission Signals

Early detection of rolling-element bearings faults is essential, and acoustic emission (AE) signals are actively utilized for monitoring bearing health condition. Most existing methods for fault diagnosis comprise two steps: feature extraction and fault classification. The convolutional neural network (CNN) is a powerful deep learning technique that can perform both feature extraction and classification procedures without the need to separate these tasks into different algorithms. However, most of the known CNN architectures are used for image recognition and require a 2-D image as an input parameter. To employ CNN to resolve the problem of rolling-element bearings fault diagnosis, in the present work, the raw 1-D AE signal is transformed into a 2-D kurtogram representation. Experimental results using eight types of various bearing conditions indicate that the proposed fault diagnosis approach utilizing the kurtogram representation of the original AE signal and CNN extracts discriminative features and achieve high classification accuracy.

Alexander Prosvirin, JaeYoung Kim, Jong-Myon Kim

CHMM-Based Classification of Dynamic Textures

Classification of dynamic textures is an important and meaningful research in texture analysis. To accurately describe and classify dynamic textures, this paper proposes Continuous Hidden Markov Model (CHMM) based method. Specifically, the implicit state variable in CHMM represents the motion information of the dynamic texture with time, and the mixed Gaussian function is used to fit the observed gray value information of the texture at the spatial position. Then, a new dynamic texture sequence is assigned to the most similar category, by calculating the maximum likelihood probability generated by the trained dynamic textures CHMM models. The experimental results on the benchmark DynTex database demonstrate that CHMM is superior to the LDS based method and DHMM based method, for obtaining higher correct classification rate.

Yulong Qiao, Na Li, Yufei Wang, Wei Xi

Survey of On-Line & Block Programming Language-Scratch: On Perspective of Educational Achievements

The objectives of computer education had been transited from general information literacy to problem-solving capabilities. Therefore, the programming and software education is more getting important and critical for computational thinking . So far, many researches verified that Scratch programming is very useful subject to improve the problem-solving capability and computational thinking skills. In this paper, we analyzed the previous researches in perspectives of academic achievements. And we suggested further research directions based on their limitations.

Jeong Ah Kim, Dae Young Ko

A Study on Lightweight Mutual Authentication Protocol Based on Simple Operation

In recent years, RFID technology has been widely applied to various fields in everyday life, and many research on RFID application technology are continuously being carried out in domestically and internationally. However, RFID technology has various security threats, and various security technologies are needed to solve them. One of them is mutual authentication between RFID components. Until now, various mutual authentication scheme have been studied in RFID systems. Especially a lightweight mutual authentication protocol suitable for a low-cost tag environment has been proposed. In this paper, we propose a lightweight RFID mutual authentication protocol that compares and analyzes existing RFID mutual authentication scheme, satisfies various security requirements, and reduces computation.

Yong-Woon Hwang, Im-Yeong Lee

Performance Analysis of Spark-DLF: Spark Based Distributed Deep Learning Framework for Article Headline Generation

In recent years, deep learning models have achieved outstanding results in computer vision and speech recognition. And now deep learning are effectively being exploited to address some major issues of Big Data, including fast information retrieval, language translation, data classification and so on. In this work, we implemented news article headline generation application for performance analysis of our framework, Spark-DLF. Training deep learning models requires extensive data and computation. Our proposed framework can accelerate the training time by distributing the model replicas, via stochastic gradient descent, among cluster nodes. We conducted a performance analysis to see how well our framework can handle Big Data problems.

Akhmedov Khumoyun, Yun Cui, Myoungjin Kim, Lee Hanku

Robust Cartoon Zero-Watermark Algorithm Based on NSST

To better extract the edge feature of the cartoon image, a zero-watermark algorithm based on non-subsampled shearlet transform is purposed. In this method, NSST can extract the edge contour feature of cartoon image well by extracting low-frequency part of the image, and the image features can be better extracted by combining normalization, DCT and non-negative matrix factorization (NMF). For the watermark image, the spread spectrum based on complete complementary codes and chaotic scrambling is used to improve the robustness and security. Finally, the zero-watermark is derived by making the feature information of the cartoon image XOR with the watermark information. The experimental results show that the method is strong robustness against noise, geometric transformation, JPEG compression.

YuChao Sun, De Li

Robust Animation Zero Watermarking Based on Visual Cryptography and Complete Complementary Code

In this paper, the experiment use common animation fragments of network as the original vector video. Firstly, we randomly select some frames from the vector video frames to embed the watermarking information. Secondly, the 2D DWT and pseudo 3D DCT transform is chosen as the processing mode to process vector video frames in the processing of transform domain. In addition, using the log polar coordinate transformation and singular value decomposition process vector video to get a binary sequence. Thirdly, in order to enhance the robustness and security of zero watermarking, we use the complete complementary code to spread spectrum the watermarking signal, and use the visual cryptography to encrypt and decompose the secret watermarking image under chaotic scrambling. Finally, in order to verify the robustness and feasibility of algorithm, we carry on the contrast experiment of two algorithms. It shows that this paper algorithm is improved against Gaussian noise, filtering and frame averaging attacks.

De Li, LiHua Cui

Secure Multicast Using Proxy Re-encryption in IoT Environment

It appears that interest in the Internet of things (IoT) has recently reached its peak, with a great deal of focus from both the private and public sectors. IoT, a technology that enables the exchange of data through linkage among all objects surrounding the user, can create new services. Data communication among objects is not limited to personal information, but can also deliver different data types, such as sensing information collected from the surrounding environment. When such data is collected and used maliciously by an attacker, it is more vulnerable to threats than in conventional network environments. Security of all data transmitted in the IoT environment is therefore essential for preventing attacks. In IoT network composed of many sensors, group communication is required. However, it is difficult to apply the conventional cipher algorithm to IoT devices. In this paper, we propose a method to forward each group key to a group member in a decentralized way using proxy re-encryption. The proposed scheme is best suited for large-scale dynamic IoT networks without a central network controller.

SuHyun Kim, ImYeong Lee

Fast Provisioning Service in Shared-Storage Based VDI According to User’s Type

The main purpose of VDI (Virtual Desktop Infrastructure) provisioning is to provide the virtual desktop and the resources that the customer wants to use quickly. The customer of DaaS appears in various forms according to the type of user. The most commonly categorized approach can be broken down depending on whether the user permanently owns the virtual desktop or whether the virtual desktop is allowed for temporary use. And the type of user will be different according to the provided policy. In this paper, we provide a mechanism to allocate virtual desktops to users in 320 ms period of time according to the form of various users, and provide a fast provisioning method by applying them to a shared-storage based DaaS system.

Dae-Won Kim, Soo-Cheol Oh, Jae-Geun Cha, Ji-Hyeok Choi, Sun-Wook Kim, Seong-Woon Kim

Design of Smart Cattle Shed System Based on BLE Beacon to Improve Power Consumption

In recent years, a cattle shed management system using the IoT technology has been introduced to resolve the reduction in productivity due to aging society and lack of labor force in the livestock industry in Korea. However, existing cattle shed management systems have not been suitable for small and medium-sized cattle sheds due to a short communication range and excessive power consumption as a result of the use of ZigBee-based wireless communication. Thus, the present paper designed a smart cattle shed system based on BLE(Bluetooth Low Energy) Beacon to improve power consumption.

Seung-Su Yang, Young-Hwan Jang, Yong-Wan Ju, Seok-Cheon Park

Design of TensorFlow-Based Proactive Smart Home Managers

In recent years, with IoT(Internet of Things) technology as the main focus, device operation and control technology in smart homes has been attracting considerable attention, and home IoT device management services are being provided by various companies, including communication companies. The smart home manager system manages smart devices used in homes, and it provides only the status value information and control function of the currently registered devices. Thus, unnecessary access procedures occur due to the characteristic of the smart home, which uses a smart device repeatedly for the same purpose. To resolve such shortcomings, in this paper, the Proactive Smart Home Manager has been designed, which can predict and suggest users the next steps to take by user usage pattern analysis and inference via machine learning.

Min-Hyung Park, Young-Hwan Jang, Yong-Wan Ju, Seok-Cheon Park

Design of Effective Indexing Technique in Hadoop-Based Database

The recent rapid increase in the amount of data to be processed has led to the increased use of dispersed parallel processing of large-scale data analysis using open-source Hadoop’s MapReduce framework. The large-data processing method proposed by Google and Hadoop which implemented this are representative dispersed parallel processing methods, and the data are dispersedly saved on the HDFS(Hadoop Distributed File System). Such HDFS uses its own indexing technique when it comes to searching specific values from the saved files. Techniques that use conventional index, however, leads to problems like reduced search performance by not considering update and saving index in the disc. Therefore, the paper proposes effective DB indexing technique on Hadoop-based database.

Jae-Sung Shim, Young-Hwan Jang, Yong-Wan Ju, Seok-Cheon Park

Transformation of EEG Signal for Emotion Analysis and Dataset Construction for DNN Learning

This work is to design an emotional analysis system using Deep Neural Network based on electroencephalogram data. The data are processed using high pass filtering and removing DC offset method in the proposed system. Then the preprocessed dataset is constructed to analysis the impact of input data placement on recognition performance. In the experiment, the happy and neutral dataset are used to measure the proposed approach performance. The result shows that learning data by stacking one row at a time is better than learning data matrix sequentially.

Yeahoon Kwon, Yiyan Nan, Shin-Dug Kim

Impact of Passive UHF RFID Reader Antenna Locations for Immobile Object Localization

In a smart environment, accurate locations of objects are a fundamental and critical issue. To achieve this goal, we present several methods based on passive far-field UHF RFID technologies, which can satisfy accuracy, robustness and reliability, cost efficiency, simplicity, compatibility, and scalability. Passive UHF RFID system based localization is easily applied to the supply chains. In this research, we studied an effect of a passive UHF RFID reader antenna’s position to immobile object localization system in the monitoring area. When the localization system uses stationary RFID reader system, the performances of the system are significantly varied depending on the deployed antenna locations due to its external environmental influences such as RF reflection by the ground and obstacles. In this research, we deeply study the RF conditions and differentiation with various antenna location.

Jae Sung Choi

Development of 3D Surface Shape Analysis System Using White Light Scanning Interference

Due to the rapid development of the industry in recent years, there is a growing demand for precise machine of parts. It is also necessary to analyze the processed products precisely in order to efficiently operate the machines in the production plant. In order to guarantee the precise of the part production, it must be possible to distinguish microscopic differences that cannot be judged by human eyes. In previous studies, only two-dimensional analysis was possible, and the analyzable system was PCB circuit eccentricity, width, diameter of circle. However, for production of precise products, it is difficult to analyze the product in two dimensions. Therefore, in this paper, we propose an extended analytical system that adds three - dimensional shape to the product so that the accuracy of the product can be clearly judged in the 2D method by providing the three - dimensional shape of the product and realizing the actual sectional shape through 3D.

Yong-Tae Jeon, Hyun Lee, Jae Sung Choi

Life Log Collection and Analysis System Using Mobile Device

With the increase in the number of smartphone users and the penetration rate now exceeding 85%, providing personalized services to individuals with smartphones has become an important research subject. In this paper, we propose a system that collects and analyzes a user’s information based on a multi-modal sensor in the mobile device to generate meaningful information based on a user’s life-log. To extract the point of interest, the system scenario was created and the experimental and analysis results were applied. We compared the performance of the classification algorithm using the evaluation scale and applied the Naive Bayes algorithm, which was judged to be an efficient algorithm, to the activity recognition system. Using this system, information can be collected and analyzed, and meaningful information can be generated and presented to the user. The information managed by this system will be used for research on services/products that are optimized for personal preferences.

Yunjin Nam, Dongkyoo Shin, Dongil Shin

Introduction of an Interactive Growing Robot/Toy for Babies

In this paper, we introduce an interactive robot/toy for babies, whose name is Buddy. As Buddy is a product for babies, its exterior is covered with combination of soft silicone and cotton for safety as well as ease physical interaction such as hugging with babies. Buddy is featured with its growing function, to accomplish true and deep interaction with babies. We designed Buddy to grow both physically and intellectually with babies, so that the babies can regard Buddy as their family or friendly peers. In addition, as Buddy will always be with babies, Buddy will act as a bodyguard as well as a keeper of valuable moments. We expect that Buddy will be a solution for issues associated with lonely babies resulted from social problems such as low fertility rate.

Jiyong Kim, Hyunsu Jeong, Azure Pham, Changhyeon Lee, Thiha Soe, Pilwoo Lee, Mingu Lee, Seong-Woo Kim, Juhyun Eune

Multiclass Data Classification Using Multinomial Logistic Gaussian Process Model

We propose the multiclass data classification method using Bayesian logistic Gaussian process model. First, we have defined the multinomial logistic Gaussian process classification model. Second, we have derived the predictive distribution of the classification variable corresponding to the new input data point by using a variational Bayesian inference method. Finally, in order to verify the performance of the proposed model, we conducted experiments using Iris real dataset. From the experimental results, we can see that the proposed model has achieved superior classification ability.

Wanhyun Cho, Soonyoung Park, Sangkyoon Kim

Implementation of Redundant Digital Excitation Control System Algorithm

Currently, Korean electric power system requires stability and reliability. All the systems demands a high reliability due to the characteristics of power plant. An algorithm for the redundant digital excitation control system has been implemented in this study. Excitation control system is one that maintains or controls output terminal voltage by supplying DC current to the field winding. Domestic controllers are needed as existing controllers are made in overseas so that it is difficult to maintain them properly. Thus, in this study, an algorithm for a redundant digital excitation control system was implemented with C language. The results of implementation were validated through on-site tests at ‘P’ power plant. The authors aims to assist system developers in understanding the flow of excitation system by disclosing the C-style pseudo code.

Hoon-Gi Lee, Hie-Sik Kim

Improving Top-K Contents Recommendation Performance by Considering Bandwagon Effect: Using Hadoop-Spark Framework

The study on the existing Collaborative filtering recommendation system is mainly aimed at improving the accuracy of prediction. However, in terms of actual recommendation service, it is more important that the Top-K recommendation list, which is effectively recommended to the user, is an item that the user actually likes, rather than improving the recommendation accuracy of all items. In this paper, we have developed a recommendation system that considers the psychological concept of Bandwagon Effect in order to improve the recommendation accuracy of the Top-K contents. For Big data distribution and storage, we used Hadoop and for the fast Big Data processing offering speed, we used Spark, an in-memory data processing framework for high-speed operations. As a result, the proposed model is superior to the existing model in terms of accuracy of recommendation for Top-K contents.

Suk-kyoon Kang, Kiejin Park

A Method for Replacing Protective Relay at Kanudi Power Plant and Validation

A modern electric power systems require both stability and reliability. This study involves validation of the method used for replacing the protective relays at the Kanudi power plant in Papua New Guinea. Replacement of 27 and 81 protective relays at the existing power plant was due to their failure. A new relays were installed after removing the power source of existing panel. The output section of each new relays was connected with existing relay in parallel as previous protective relay only allowed monitoring without outputting the contents so that the new protective system was designed to in a way that both existing and new relays perform detection. The validation after the replacement was successful. The new system aims to protect power generator by coping with accidents.

Hoon-Gi Lee, Hie-Sik Kim

A Critical Review on Operation and Management System of Doping Control Officer in Korea Anti-Doping Agency

The purpose of this study is to discuss doping, an issue which has been magnified by the ever critical view of an information-oriented society, and suggest a solution to this issue [1–3]. First of all, let us discuss the characteristics of modern society and the use of doping. Doping can be described as a quiet struggle between players who want to hide their use of doping and the doping control officers (DCO) who want to catch it. Currently, players are able to take advantage of an information-oriented society to evade detection. Some doubt has been expressed concern as to pretesting schedules are often leaked out ahead of time in KBO (The Korea Baseball Organization). It implies that there are security vulnerabilities via DCOs or hackers, allowing for information leakage. To resolve this, the study suggests (1) selecting full time DCOs; (2) tightening security within the DCO management system; (3) expansion of doping control and (4) ethics education for DCOs.

Seung-Hoon Lee, Chang-Hyeok Seok

The Issues of Media and Public Ethics on Doping Problems

The purpose of this study is to discuss the doping problems that can be caused by various media and SNS (Social Networking Service) which are characteristics of the information society. It is easy to contact and produce a lot of information through a variety of media and SNS. Similarly, the information doping follows the same scheme. The spread of false information about doping is a serious problem as it can lead to the spread of doping. The core of anti-doping policy lies in ‘Principle of Strict Liability’ and ‘Principle of Zero Tolerance’. This is based on Deontology, which is a hard punishment for doping offenders. However, the media and the press are often opposed to advocacy or sympathy for doping offenders. This is a problem caused by a lack of awareness of doping, which corresponds to the dysfunction of the information society. To prevent this, a pan-national understanding of anti-doping policy is needed. It is necessary to instill understanding and awareness of doping which is the ethical consciousness of the media and the public through education and promotion using information media.

Seung-Hoon Lee, Chang-Hyeok Seok

Seamless Reconfiguration of Virtual Dedicate Network Based on Software Defined Network

Recently, Software-Defined Networking (SDN) technologies have considered a nice way to realize network virtualization. SDN can efficiently and automatically manage virtual network(s) on physical network by allocating the substrate network resources. In this paper, we propose a seamless virtual dedicate network (VDN) reconfiguration method based on center node with highest closeness centrality for supporting new user demands. This method have no effect on the communications related with existing VDN, and also improves the performance of VDN reconfiguration.

Yong-hwan Kim, Ki-Hyeon Kim, Joon-Min Gil, Dongkyun Kim

Design and Implementation of Digital Doorsign System Based on E-paper Display

In this paper, we proposed digital doorsign system based on E-paper display for reducing cost and time in replacing doorsigns and effectively communicating with visitors. We also showed implementation results of the proposed system. Because the system could create final notice by only using standard forms which is designed by system manager, it is possible to keep uniformity of institution. By posting announcement for temporary event venues, users could communicate with visitors. We also designed the system which uses E-paper display and BLE for low power consumption in mobile device, Digital Doorsign.

Bong-Ki Son, Jaeho Lee

A Novel for Light-Weighted Indoor Positioning Algorithm with Hybridizing Trilateration and Fingerprinting Method Considering Bluetooth Low Energy Environment

IPS (Indoor Positioning System) has been considered as one of mostly challengeable technical issues at the focus of accuracy and differentiation even though it has been researched long times ago, and many of researches found that fingerprinting approach can show better accuracy comparing with previous trilateration-based schemes. However, regarding huge indoor place, there can be significant calculation problem issue due to a lot of cells to be pre-surveyed radio map and the great number of location anchor point such as beacon device. Furthermore, this can aggregate life time of user handheld devices which would be generally tiny shape and insufficient energy resource. In this paper, we present a new scheme which was researched with a unique goal for improving calculational computing power, with hibridizing traditional trilateration and recently issued fingerprinting approaches under the Bluetooth low energy environment. In addition, we also present experiment results with analytical evaluations to prove the performance of proposed scheme.

Jaeho Lee, Bong-Ki Son

Design of Beacon-Based Positioning System Using RF and Sound Wave in Smartphone

Along with development of IoT (Internet of Things), the importance of indoor location-based services is increases. Therefore, researches on beacon, which is widely used for indoor positioning, have been actively implemented. Among them, location positioning using Beacon’s RSSI (Received Signal Strength Indication), which heavily used, but it is less accurate because it relies on the distance based on the signal strength. In this paper, Beacon transmits RF signal and specific frequency sound wave. The smartphone receives the two signals and converts them into distance values through the Time Difference of Arrival (TDoA) method. We propose a position location system that uses calculated distance values for trilateration in a smartphone. It is possible high accuracy even in existing smart phones.

Hyun-Seong Lee, Seoung-Hyeon Lee, Jae-Gwang Lee, Jae-Kwang Lee

A Study on the Designing of a Laboratory Accident Cognition Model Using Smart Sensor Based Decision Tree

There are diverse risk factor in the laboratories of universities and research institutions according to research purposes, accidents caused by this are highly likely to accompany secondary damages such as fire, explosion. Presently, sensors are utilized to manage safety. They are only capable of supporting the decision making process, but have yet to able to cognize and process accidents on its own. For this reason, the study researches a model capable of cognizing, analyzing of accident autonomously and informing users, based on smart sensors. To this end, designs a system that utilizes smart sensor and Arduino to collect factors such as temperature, transmits them to the main server, and save. Then, in order to cognize accidents, use the decision tree algorithm at the main server to find the threshold value indicating accident occurrence. By doing this, presents a model designing capable of preventing secondary damage expansion and reducing losses.

Ki-Su Yoon, Seoung-Hyeon Lee, Jae-Pil Lee, Jae-Kwng Lee

Positioning Model Design Using Beacon and Geomagnetic Sensor of Smartphone

As the indoor location service is growing, its related studies based on ZigBee, UWB, or Beacon are actively done. Indoor location requires higher accuracy than that of outdoor since the traveling distance is relatively shorter. But accurate user location is not easy to acquire due to such problems as fading, inter-floor interference or sensitivity of the signal receiver. Therefore, we are using the triangulation by RSSI of BLE Beacon to establish the initial user location and the Geomagnetic Sensor based PDR of Smartphone to determine the user location in this study.

Jae-Gwang Lee, Seoung-Hyeon Lee, Jae-Kwang Lee

A Design of Scheduling Program for Diabetic Patients: A Software Engineering Approach

Today, the Republic of Korea (ROK) is battling with diabetes which is being increasingly found not only in the elderly but also in young people due to the westernized eating habits. Hyperglycemia being its main cause, this disease is closely associated with one’s lifestyle such that it is the most important task for diabetic patients to improve their lifestyles to recover from this disease or improve its symptoms. This is the study of a convenience program for improving the lifestyles of diabetic patients from the software engineering point of view. The designed software allows the patient to recognize the relationship between sugar level and treatment plan and recommends an adequate treatment measure by providing relevant information through push notification service in accordance with the patient’s condition. This study aims to assist patients to prevent or treat diabetes by themselves and change their lifestyles properly.

Jeong-Hoon Choi, Jun-Ho Huh, Sunghyun Weon

A Simulated Infiltration Test for Network in Virtual Environment Using VMware Virtualization Technique

Due to rapid development of IT technology, almost all things are being databased and as the value of information assets, the number of cyber attacks is increasing. Accordingly, the role of the white-hackers is becoming important, as well as their training. For the training, a simulated network infiltration test environment has been constructed in this study by using VMware’s virtualization technique to allow potential white-hackers to study the basics in networking area and some of the major attacking techniques which were actually put into the tests.

Guk-IL Kim, Jonghyeon Kim

Development of Network Cyber Attack Training Curriculum for Security Staff

Recently, the frequency of cyber attacks targeting the information assets of firms or institutions is continuously increasing. Although there are measures such as reinforcement of security policies, regulations, information protection organizations and physical security in addition to replacement of security equipments for protecting the assets, the core of the cyber security is responding to the security breaches. For this reason, producing well-trained security personnel (computer emergency responders) is essential. Although there are many publications dealing with the theories of cyber security, the learning materials that can be referred by the security personnel mainly for practices are somehow in short supply. Thus, in this study, a curriculum for network cyber attack training has been developed for the security personnel. The curriculum focuses on actual implementation of network-attacking environments, simulated attacking methods, attack pattern analysis and countermeasures.

Guk-IL Kim, Jonghyeon Kim

A Design of Portable Continuous Passive Joint Mobilization Equipment System

As for the integration with IT in the medical field, an attempt to introduce some systems integrated with the equipments having a network capability like smartphones connected to medical equipments is not rare. The proposed design in this study aims to supplement the disadvantage of existing Continuous Passive Motion (CPM) and develop a system that can treat and manage patients by enhancing portability and usability through integration with IT devices. The device design is largely divided into the main body comprising of (1) axis part, (2) fixed arm part and (3) moving arm part and the operational part (an exclusive application). As two electric motors that are necessary to operate the machine are attached on either side of the axis part in the center, flexion and extension movements are possible. The second part, fixed arm, plays the role of anchoring a part of human body while the moving arm part is applied to the place where movement is actually necessary. The design allows the system to be used for the treatment of patients anytime and anywhere. Also, as it can operate the equipment and manage treatment information through its exclusive application the equipment application can be maximized. Thus, it’s been judged that the proposed system will be able to replace existing CPM and ultimately improve the quality of human life with its integrated form.

Do Yeon Jeon, Young Hyo Kim, Ha Yeon Park, Jun-Ho Huh, Hyeok Gyu Kwon

Delivering Historical Information Shown in Korean TV Drama Over Smart TV

We will analyze that some historical dramas which are motivated by historical and cultural contents in Joseon Dynasty. Based on this analysis, I suggest efficient service model to provide better watching experience. In particular, I focused on how to collect historical and cultural information related to TV drama, how to reproduce such information for viewers’ own needs. Also, I consider services in several languages for foreign viewers.

Jisun Byun

An Efficient Clustering Technique for Unstructured Data Utilizing Latent Semantic Analysis

Recently, data analysis has been actively conducting. In particular, meaning in a sentence is determined by using the frequency of unstructured data such as SNS and opinions which are analyzed on a specific topic based on positive or negative words. However, this analysis involving only a simple word frequency or a specific word which includes unnecessary data. In this paper, we use the analytic technique of the referred documents, exponentialize the frequency of words. We integrate the frequency of words in exponentiated sentences with each other, and finally implement an analytical model that can quantitatively compare sentences.

Yonghoon Kim, Mokdong Chung

An Improving Algorithm of Generator Reactive Power Reserves Calculation Considering Effective Generators in IEEE-39 System

One of the major element contributing to the stability of system voltage is available reactive power so that it is vital that the system operator should always keep track of how much generator reactive power reserves are currently available. Generator reactive power reserve is usually calculated by deducting current available reactive power from the maximum reactive power that can be generated by the generator. Even so, such a calculation alone does not ensure stability of system voltage. Therefore, the sensitivity matrix has been considered in this study. This matrix is useful when identifying the correlations between generator and load buses. The effective reactive power reserve (EQR) can be estimated by using this matrix to achieve voltage stability instead of employing the conventional reactive power reserve (CQR).

Moonsung Bae, Byongjun Lee

Load Loss Coefficient and Power Loss Tracing in Power Systems

A definition of Load Loss Coefficient (LLC) is given in this study along with the power loss tracing algorithm. As LLC indicates the effect of load on power transmission loss, its calculation is performed based on the Bialek’s power tracing method, where gross and net flows are being considered, to determine the power loss in a system during power transmission.

Moonsung Bae, Byongjun Lee

A Power Generation Rescheduling Method Under Power Flow Constraints Using Power Flow Tracing

When it is necessary to adjust the power flow over a certain bus, the method that best uses given conditions to efficiently adjust power flow is essential. There are several ways to control the power flow on transmission lines by adjusting the power generations but the method that controls the flow by rescheduling the generation is not adequate for use as it is not easy to carry out power flow adjustment on a particular line or it is not clear which power generator’s power generation capacity should be adjusted as the active power has little regional constraints. Thus, a method that can cope with these problems is required. Therefore, a method that trace the origin of power flow over a particular power line by using the power flow tracing technique to control power generation level and eventually adjust power flow has been proposed in this study.

Moonsung Bae, Byongjun Lee

The Effect of the Intrinsic Quality of UI of Mobile Apps on the Behavioral Intention to Use the Apps

From the time when mobile devices such as smartphones and tablet PCs launched in the late 1990s, the user interfaces (UI) of mobile applications have become more complicated and ambiguous. Such user environments limit the ease of use of applications and negatively impact not only the intrinsic quality of the UI, but also users’ behavioral intention to use the applications. However, a literature review shows that prior research in information systems (IS) has not shown much interest in the relationship between the intrinsic quality of the UI of mobile applications and users’ behavioral intention to use said applications. The main research goals of this study are: (1) to examine the direct effects of the accuracy and reputation of the UI of mobile applications on users’ behavioral intention to use the applications, and (2) to find the relationships between the accuracy of the UI of the applications and the ease of use of the applications as well as between the ease of use of applications and the reputation of the UI of the applications. A survey was conducted to collect data, and structural equation modeling (SEM) was then used for the analysis. The results of this study indicate that the accuracy of the UI of mobile applications has a significant, direct effect on the ease of use of the applications and on users’ behavioral intention to use said applications. In addition, the ease of use of the mobile applications also has a significant effect on the reputation of the UI of the applications as well as on users’ behavioral intention to use said applications. Finally, a direct effect was also found for the reputation of the UI of the applications on users’ intention to use the applications.

Wonjin Jung

Consideration of Privacy Risk Assessment of the My Number in the Financial Industry in Japan

In Sep. 2015, the Act on the Use of Numbers to Identify a Specific Individual in the Administrative Procedure was revised. It was decided to link personal numbers to deposit numbers of financial institutions. Currently, the Privacy Impact Assessment which is obliged to implement this law is required to implement safety control measures for the private sector. However, there is no system to conduct a risk assessment of the law. In the financial industry, which is a highly private sector of public nature, some privacy risk assessment is required because it has many individual numbers. In this paper, we propose a framework for privacy risk assessment on this law in the financial industry, using the privacy impact assessment prescribed as an international standard.

Sanggyu Shin, Yoichi Seto, Kei Sakamoto, Mayumi Sasaki

VANETs-Based Intelligent Transportation Systems: An Overview

The rapid development of technologies in Vehicular Ad-hoc Networks (VANETs) field led to the creation of different systems and enriched researches by different private, public sectors and researchers around the world. These achievements provide a strong and secure communication between vehicles themselves (V2V) and vehicles with infrastructures (V2I). This paper gives an overview of most of the technologies that has been introduced in VANETs-Based Intelligent Transportation Systems up to date. We classified them based on their functionalities, including congestion avoidance, controlling the intersection, accident avoidance, and emergency management.

Sarah Baras, Iman Saeed, Hadeel A. Tabaza, Mourad Elhadef

Designing Network-Attached Storage Architecture for Small and Medium Enterprise Applications

The Network Attached Storage (NAS) embraces the fascinating features of the Cross-platform, Web Hosting, File Sharing, Cloud Service, Email, Security Control and Multimedia technologies. Currently, NAS is classified as two types, Storage NAS and Platform NAS. The research of this paper focuses on Platform NAS and realizes the mechanical properties and performance of Platform NAS to effectively develop the system architecture for Small and Medium Enterprise (SME) businesses. At present, the applications of Platform NAS technology are still not fully unveiled. The research of this paper, from the business perspective, illustrates a new business operational dimension for SME that steeply cuts down the operational costs of hardware and software required by most SME businesses. From the technical perspective, the research is to experimentally reveal the advantages of NAS in the aspects of web hosting, file sharing, cloud services, email server, certified security and surveillance controls, and multimedia technologies, which are essential to SME. The results and evaluation of the paper demonstrate the real world and practical business cases under the NAS server, in which it significantly shows the contribution of the software architecture development, the amortization of the affordable business costs, and the provision of the reasonable and flexible operational security for SME sectors in the commercial cities in developing countries such as Hong Kong in Asia.

Andy Shui-Yu Lai, Anson Man-Sing Ma

Limited Constraint Problems Reasoning on Computer Based on “And/Or” Tree

In traditional ways, we deal with the reasoning on computer in many ways, such as OWA, fuzzy Petri nets or etc. But all of these methods could not apply to solve the constraint problems. This paper gives some definitions about the constraint problems and propose a solution which is based on the “and/or” tree structure to solve the limited constraint problems. It can be more flexible and efficient. At the end of this paper, there is a specific and actual example which is an expert system that uses this solution to show how it works. It proves that this method could be applied on solving limited constraint problems.

Danyang Cao, Lina Duan, Xue Gao, Lei Gao

Impact of the Selection of a Communication Tool on the Effectiveness of Media Outputs About the Results of Science and Research

Currently, media have a significant impact on the provision of information for population. They can have both a positive and negative opinion effect on different kinds of activities, including those of scientific and research institutions. The purpose of this article is to examine the impact of the selection of a communication tool on the effectiveness of media outputs about the results of science and research. This was done with the methods of a literature review and a field observation. The findings indicate that the most effective tool seems to be a personal lobbying - direct meetings with the journalists. On the contrary, the least effective tool is then a press release.

Zuzana Bouckova, Blanka Klimova

TCP Performance of CSMA/CA Wireless Networks in Different MCS and PPP Settings

The popularity of CSMA/CA based wireless networks such as IEEE 802.11 standards has resulted in deployment of many wireless network zones in various areas such as residences and public spaces. To analyze and improve the performance of these wireless networks, many researchers recently have adopted methods from stochastic geometry theory. However, for more realistic analysis of wireless networks, we think elastic Internet flows such as TCP traffic need to be considered along with stochastic geometry setups. In this paper, we use Poisson point process to model the geographic distribution of IEEE 802.11a access points in given areas and evaluate TCP performance in various scenarios such as different AP densities and MCS levels through simulations.

Soohyun Cho

Expected Patch Log Likelihood Based on Multi-layer Prior Information Learning

How to preserve the edge and texture details has been a difficult problem in image denoising. In this paper, we propose a multi-layer prior information learning method, which combines the statistical and geometric features of the image to describe the attributes of the prior information more accurately and completely. The experimental results show that our proposed method is superior to the EPLL (Expected patch log likelihood) method with a single statistical characteristic for a priori learning in both visual and quantitative evaluation.

ShunFeng Wang, JiaCen Xie, YuHui Zheng, Tao Jiang, ShuHang Xue

Decentralized E-Voting Systems Based on the Blockchain Technology

This research is aimed to design a decentralized e-voting system. The core idea is to combine the blockchain technology with secret sharing scheme and homomorphic encryption in order to realize the decentralized e-voting application without a trusted third party. It provides a public and transparent voting process while protecting the anonymity of voter’s identity, the privacy of data transmission and verifiability of ballots during the billing phase.

Jen-Ho Hsiao, Raylin Tso, Chien-Ming Chen, Mu-En Wu

Reputation-Based Trust Evaluation Mechanism for Decentralized Environments and Its Applications Based on Smart Contracts

The greatest feature of the blockchain is the decentralization. This paper refers to the PGP web of trust and theoretical concepts of six degrees of separation to establish a set of reputation-based trust evaluation mechanism for decentralized environments. It is expected that achievements of the paper can facilitate people’s judgment regarding the reliability of strangers and reduce the risks of being deceived.

Kun-Tai Chan, Raylin Tso, Chien-Ming Chen, Mu-En Wu

Password-Based Authenticated Key Exchange from Lattices for Client/Server Model

We proposes a password-based authenticated key exchange from lattices for Client/Server model. The client only has to remember the password shared with the server, and the server records the password in addition to its own public/private key pair. Both parties execute the mutual authentication via the shared password and accomplish the key exchange within two steps. The security of our protocol is based on the LWE problem of lattices, so it is secure even an attacker uses a quantum computer.

Yi-Siou Jheng, Raylin Tso, Chien-Ming Chen, Mu-En Wu

Ocean Adventure Cell Phone Adventure Game Design and Implementation

In recent years, cell phone game industry continues to develop in China, and the type of games has been continuously enriched. But in a wide variety of cell phone games, ocean-themed adventure game is not much. Topic of this paper is different from other thrilling adventure games, while making a fresh ocean-themed adventure cell phone game. We complete the dynamic game background, human-computer interaction, scene interaction, rank list, social sharing and audio effects in the game. We also use the Unity3D engine to load resource, control interactive, collision detection, and data management. In the end, we transplant the game to Android platform, and the game realizes the basic functions and runs smoothly.

Xingquan Cai, Chao Chen

Analysis of Thermocline Influencing Factors Based on Decision Tree Methods

Natural phenomena disturb marine ordinary states mainly by disturb the sea surface temperature. As temperature is the main factor that affects thermocline, in this paper we propose a method to quantitative analyze the correlation between El Niño and thermocline based on decision tree methods rather than qualitative analysis. The experiments use the refined BOA_Argo data and the decision trees are constructed with these data. We aim at making better use of thermocline and trying not to be harmed by natural disasters such as El Niño.

Chengquan Hu, Yu Gou, Tong Zhang, Kai Wang, Lili He, Yu Jiang

Thermocline Analysis Based on Entropy Value Methods

Temperature, salinity, and geographic locations are three important factors while determining thermocline. We mainly focus on analyzing how these factors affect the formation of thermocline using machine learning methods. An improvement based on ‘entropy value method’ while choosing thermocline is demonstrated in the paper. The experiments adopt Argo data sets and the experimental results show that machine learning methods can compute thermocline and related data effectively.

Chengquan Hu, Yu Gou, Tong Zhang, Kai Wang, Lili He, Yu Jiang

Design and Implement of Intelligent Insole System

Wearable computing will have effect our life deeply by 2025. Up to now, there is no pressure detecting insole designed according to the anatomical partition. In this paper, we design an intelligent insole, which is divided into 9 zones. This intelligent insole can collect and record user’s plantar pressure for each zone and display data on the mobile phone in the form of charts, curves and figures. We compare the obtained data with the healthy state model, then gives some hints and suggestions base on analysis, that are propitious to user health. The intelligent insole can infer the possible diseases and give the suggestion of convalescence. Experimental results show that intelligent insoles have good performance, which is low cost and easy popularization.

Ruisheng Li, Bin Ma, Lili He, Jin Wang

Fitness Sport Data Recording System Design and Implementation on Smart Phone

With the rapid development of the smart phones, the advantage of using smart phones to help assist with daily activities is obvious; thereinto, using smart phones to record sport data is a prevalent trend. According to this direction, this paper designs and implements the fitness sport data recording system which contraposes the fitness sport data rather than common sport data on smart phone. This paper designs and implements three main modules respectively are user-defined fitness sport data adding, complex fitness sport data exhibiting, complex fitness sport data classified summarizing and eventually accomplishes the system. The practical operation effect expresses that the fitness sport data recording system designed by this paper runs well.

Xingquan Cai, Runbo Cai

Construction of High Resolution Thermocline Grid Data Sets

Thermocline has always been the emphasis of marine research. In this paper, we propose a method to construct high resolution marine grid data sets on the basis of MLP. Data used in the article is from World Ocean Atlas 2013. The experiments show that high resolution data can calculate the depth, thickness and strength of thermocline precisely. The method is vital to thermocline gridding.

Chengquan Hu, Tong Zhang, Jin Wang, Yu Gou, Kai Wang, Hongtao Bai, Yu Jiang

Service Identification Framework for Systems of Systems Based on MPLS Technology

Systems of Systems (SoS) are collections of autonomous systems that share their resources in order to form a new larger system with more capabilities and enhanced performance. A fundamental challenge of service identification in SoS arises when constituent systems are dynamically composed. This paper proposes a new framework for SoS service identification and allocation based on MPLS technology.

Sahel Alouneh, Dhiah el Diehn I. Abou-Tair, Ala Khalifeh, Roman Obermaisser

FPGA Based Face Detection Using Local Ternary Pattern Under Variant Illumination Condition

This paper presents the design and implementation of real-time face detection using Local Ternary Pattern (LTP). First, an input image is transferred by the Camlink interface and the image is then downscaled for face detection. A tree-structured cascade of classifiers is used for face detection. We implemented the proposed hardware architecture on a Xilinx Virtex-7 FPGA and the processing speed was adjusted to the frame rate of the camera. The size of the input images is 640 × 480 (VGA) and a larger size can be used without performance loss.

Jin Young Byun, Jae Wook Jeon

Finding Similar Microblogs According to Their Word Similarities and Semantic Similarities

Weibo is the biggest microblogging service in China, whose users publish a mass of short messages every day. Whenever there is a hot event, weibo users will scramble to publishing the messages about this event and most of these messages are repeated. In this paper, we proposed two different methods to identify similar weibo messages from the massive candidate sets. The first method is based on the “Simhash”, who can find the similar messages based on their word similarities. The second method is based on the “Paragraph Vector”, which identify the similar messages based on their semantic similarities. We collect a real-world dataset to conduct experiments, and the results show that our two methods can efficiently identify the similar messages.

Yuan Wang

An Improved Trust Model Based on Time Effect

Trust model is a very important model in social networking and recommendation technology. The trust relationship between the users can reflect the relation in the real life in a better performance than the similarity, but this kind of trust model lose sight of the importance of time effect, so this article pay attention to the research on time effect in trust model and proposed an improved trust model based on time effect. At last we choose appropriate data set to prove the superiority of the proposed improved trust model.

Zhichao Yin, Hui Zhang, Chunyong Yin, Jin Wang

Text Classification Algorithm Based on SLAS-C

Nowadays, mobile marketing is becoming increasingly important both strategically and economically because of the mobile devices. Short text is becoming a popular text form which can be seen in many fields such as network news, QQ messages, comments in BBS and so forth. Besides, our mobile devices also contain a lot of data of short text. To extract useful information from the short text more efficiently, this paper proposes SLAS (semi-supervised learning method and SVM classifier) and CART (classification and regression tree) to improve the traditional methods, which can classify massive short texts to mining the useful information from the short texts. The experiment also shows a better result than before, which has a more than 10% increase, including precision rate, recall rate and F1 value, besides, the running time is reduced by half than the KNN algorithm.

Zhichao Yin, Jun Xiang, Chunyong Yin, Jin Wang

A Ground Segmentation Method Based on Gradient Fields for 3D Point Clouds

In order to navigate in an unknown environment, autonomous robots must distinguish traversable ground regions from impassible obstacles. Thus, ground segmentation is a crucial step for handling this issue. This study proposes a new ground segmentation method combining of two different techniques: gradient threshold segmentation and mean height evaluation. Ground regions near the center of the sensor are segmented using the gradient threshold technique, while sparse regions are segmented using mean height evaluation. The main contribution of this study is a new ground segmentation algorithm that can be applied to various 3D point clouds. The processing time is acceptable and allows real-time processing of sensor data.

Hoang Vu, Hieu Trong Nguyen, Phuong Chu, Seoungjae Cho, Kyungeun Cho

Recognizing the Adhesion Hollow Characters Based on the Closed Cutting Algorithm

With the development of network security, CAPTCHA identification is becoming a hot research topic. In this paper, we study the adhesion hollow character in the CAPTCHA, and put forward a method to identify the hollow characters which are adhered to each other in the CAPTCHA. The method is based on the principle of virus invading cells, and uses the particle swarm algorithm to find the closed space and fill it. We determine the location of the character by the histogram of number of layers, and then recognize the characters that are cut out. Experimental results show that the proposed method can be effectively used to identify the hollow characters in the CAPTCHA, and the recognition rate is 30.1%.

Chunyong Yin, Kaiwen Zhu, Lian Xia, Jin Wang

Ensemble R-FCN for Object Detection

This paper presents an Ensemble R-FCN framework for object detection. Specifically, we mainly make three contributions to our detection framework: (1) we augment the training images for R-FCN when facing the limited training samples and small object. (2) We further introduce several enhancement schemes to improve the performance of the single R-FCN. (3) An ensemble R-FCN is proposed to make our detection system more robust by combining different feature extractors and multi-scale inference. Experimental results demonstrate the advantages of the proposed method. Especially, our method achieved the performance of AP score 0.829 which ranked No. 1 among over 360 teams in Ucar Self-driving deep learning Competition.

Jian Li, Jianjun Qian, Yuhui Zheng

A Design of Evolutionary Personal Information Partner Based on Software as a Service

We are now having to face massive amounts of data, and need an autonomous, active, capable information system to help us for getting the useful data. To satisfy the requirement, an autonomous information system which is called Evolutionary Personal Information Partner has been proposed. We have proposed the mechanism of autonomous evolution, its effectiveness has been confirmed by using a case study. But the design of Evolutionary Personal Information Partner has not been proposed until now. On the other hand, Software as a Service has become a common delivery model for many business applications. This paper proposes a design of Evolutionary Personal Information Partner based on Software as a Service.

Yifeng Han, Hongbiao Gao, Jingde Cheng

A New Method of Arm Motion Detection Based on MEMS Sensor

This paper proposed a new arm motion detection method based on a MEMS sensor. The method gets three-dimensional accelerations and angular velocities of user’s arm motions from a MEMS sensor. Then calculate the correlation coefficients between the data of the current motion cycle and each set of data in the feature database after normalize the data. Thereby, the system can successfully detect 2 specific motions, and the accuracy rate of the detection is 90%.

Kai Wang, Chengquan Hu, Lili He, Fenglin Wei, Yu Jiang

Fitness Device Based on MEMS Sensor

Nowadays, motion detection technology is an important field of investigation especially for those researchers whose field is human-computer interaction. Visual algorithms are generally getting complicated when the scale of information is huge. Under most of the situations, calculations need to be done rapidity. Vision sensor may not that appropriate. MEMS provides low dimensional data with stronger adaptability for various occasions. This paper represents a fitness device in which an acceleration sensor can capture users’ movements. Experimental results confirm the feasibility of the fitness devices.

Fenglin Wei, Chengquan Hu, Lili He, Kai Wang, Yu Jiang

A Security Framework for Systems-of-Systems

Systems of systems have been used recently in many complex scenarios to simplify network communication, resources allocation and management. Given the dynamic and heterogeneous nature of systems of systems, providing security in such an environment is challenging and requires significant management and processing resources. In this paper, a security framework based on the Multi-Label Switching Protocol is proposed, which not only provides the network with connectivity, reliability and Quality of Service, but also adds several security features such as traffic separation and isolation with minimal management and configuration overhead. Furthermore, a more advanced security configuration based on the integration of IPsec and Multi-Label Switching Protocol is proposed.

Dhiah el Diehn I. Abou-Tair, Sahel Alouneh, Ala Khalifeh, Roman Obermaisser

A PSO Based Coverage Hole Patching Scheme for WSNs

In this paper, a mobile assist node deployment algorithm is proposed to patch coverage holes in wireless sensor network (WSNs). In initial phase, sensor nodes are randomly deployed in target area, they remain static or switch to sleep mode after deployment. Then, we partition the network into girds and calculate the coverage rate of each gird. Finally, we wake mobile sensors from sleep mode to fix coverage hole, and we utilize particle swarm optimization (PSO) algorithm to calculate the moving position of mobile sensors. Simulation results show that our algorithm can effectively improve the coverage rate of WSNs.

Jin Wang, Chunwei Ju, Hye-jin Kim, R. Simon Sherratt, Sungyoung Lee

A Novel Feature-Based Text Classification Improving the Accuracy of Twitter Sentiment Analysis

With the growth of Internet and various online services, tremendous amount of data are generated in real time. As a result, sentiment analysis of online reviews has become an important research problem. In this paper a novel feature selection and weighting scheme is proposed for the sentiment analysis of twitter data. The Part of Speech (POS) tagging and Bayes-based Classifier are utilized in the proposed scheme. Also, different from the existing schemes, independency of the attributes and the influence of emotional words are properly manipulated in deciding the polarity of test data. Computer simulation with Sentiment 140 workload shows that the proposed scheme significantly outperforms the existing sentiment analysis schemes such as naïve Bayes classifier and selective Bayes classifier.

Yili Wang, Le Sun, Jin Wang, Yuhui Zheng, Hee Yong Youn

System Information Comparison and Analysis Technology for Cyber Attacks

Although info-communication technologies are improving the quality of life, the damage due to insufficient security polices and various and powerful hacking techniques are increasing, affecting the national infrastructure. In this paper, we describe a technique to integrate digital forensic technology and an attack tree when a cyber attack is performed on any system, and to reveal the method and path of the attack through the storage, comparison, and analysis of a system’s volatile information based on the time difference. The proposed technology is expected to help assess the damage to the institutional systems networks, which will therefore allow solutions to be quickly determined.

Hyeonsu Youn, Duhoe Kim, Yong-Hyun Kim, Dongkyoo Shin, Dongil Shin

A Study on Comparison of KDD CUP 99 and NSL-KDD Using Artificial Neural Network

Computer network face many security problems because various smart devices using computer networks are being developed and are rapidly spreading. Therefore, an Intrusion Detection System (IDS) is necessary for network security. There are two typical datasets for IDS: KDD CUP 99 (KDD’99) and NSL-KDD. In this paper, we introduced KDD’99 and NSL-KDD and analysis these datasets using an artificial neural network. KDD’99 shows a higher score for overall accuracy, but falls behind in its classification accuracy per category.

Hyunjung Ji, Donghwa Kim, Dongkyoo Shin, Dongil Shin

The Application of Augmented Reality Based on Body Parts Recognition

Augmented reality is through the integration of computer generated virtual images and real environment to let the user access to feel real with rich content experience and human-computer interaction technology. Based on the augmented reality and human body recognition technology, the research builds an interactive augmented reality application of propaganda enterprise culture. Through the test in practical application and comparing the method based on color image recognition of human body parts and the method based on depth image recognition of human body parts, this paper proposes a relevant method of improving the recognition performance.

JianFeng Hou, Wei Song, MengXuan Li

A Novel Subspace Super-Pixel Based Low Rank Representation Method for Hyperspectral Denoising

This paper presents a novel denoising method based on subspace superpixel based low rank representation for hyperspectral imagery. First, the original hyperspectral data is assumed to be low-rank in both spectral and spatial domains. The spectral low rankness of HSI data is represented by decomposing it into two sub-matrices of lower rank while the spatial low rankness is explored within superpixel based regions in the subspace. The superpixels are generated by utilizing state-of-the-art superpixel segmentation algorithms in the first principle component of the original HSI. The final model could be efficiently solved by augmented Lagrangian method (ALM). Experimental results on simulated hyperspectral dataset validate that the proposed method produces superior performance than other state-of-the-art denoising methods in terms of quantitative assessment and visual quality.

Le Sun, Yili Wang, Jin Wang, Yuhui Zheng

A Chinese Handwriting Word Segmentation Method via Faster R-CNN

The segmentation of Chinese handwritten document image into individual words is an essential step for the character recognition. Conventional methods frequently use feature extraction and classification algorithm to segment. However, since the features of the words mostly depend on people, it is considered a difficult task. In order to avoid this problem, we use a method of object detection—Faster R-CNN. The words are treated as the especial object and people do not concern on features extraction. Experimental results on HIT-MW databases show that our method achieves the preferable performance.

Zelun Zhang, Jin Liu, Chenkai Gu

Aspect Based Sentiment Analysis for Online Reviews

Learning good semantic vector representations for sentiment analysis in phrases, sentences and paragraphs is a challenging and ongoing area of natural language processing. In this paper, we propose a Convolution Neural Network for aspect level sentiment classification. Our model first builds a convolution neural network model to aspect extraction. Afterwards, we used a sequence labeling approach with Conditional Random Fields for the opinion target detection. Finally, we concatenate an aspect vector with every word embedding and apply a convolution neural network over it to determine the sentiment towards an aspect. Results of an experiment show that our method performs comparably well on Yelp reviews.

Lamei Xu, Jin Liu, Lina Wang, Chunyong Yin

Chinese Anaphora Resolution Based on Adaptive Forest

Anaphora resolution is one of the key problems in natural language processing. In natural language, in order to make language concise and to reduce redundancy, often using different words to replace the words or sentence of the same meaning. However, it is difficult for a computer to understand these issues as a human. Some researcher proposed using decision trees to solve this problem, but decision trees may have problems with over-matching. In this paper, we provide a better way called adaptive forest which combine random forest and adaptive boosting to resolve this problem. Experiment result shows the effectiveness of our method in anaphora resolution.

Yunqing Zhao, Jin Liu, Chunyong Yin

Construction and Application of Fuzzy Ontology

Nowadays, there are a lot of ontology construction methods, but they cannot express ontology clearly. In this paper, we compare the existing ontology construction methods and then summarize the advantages and disadvantages of these methods. Unlike previous work, we focus our attention on the fuzzy ontology and propose a semi-automatic method for constructing fuzzy domain ontology. Then we apply it to guide the crawler in sorting the web pages in the shipping field. The experimental results show that our method has a better effect on the ranking of web pages.

Li Lin, Jin Liu, Yuhui Zheng

Generating Realistic Chinese Handwriting Characters via Deep Convolutional Generative Adversarial Networks

A person can hardly write a totally same handwriting character, more or less, there will be some tiny difference between each character. Usually, we use a neural network to generate handwriting characters, but each time we want this model to output a character, it will always the totally same. To solve this tiny different problem, we use a special neural network called DCGANs (deep convolutional generative adversarial networks). Experiments show that our method achieves good performance.

Chenkai Gu, Jin Liu, Lei Kong

Hadoop Based Parallel Deduplication Method for Web Documents

This paper proposes a method of deleting duplicate web pages through tf-idf and splay tree. According to the keywords which are extracted by TextRank, those pages which may be duplicate copies will be sent to a group. Then these pages will be judged by the method above. We use three Map-Reduce tasks to ensure the method of calculating tf-idf and deleting duplicate web pages. The experiment result shows that the algorithm can remove duplicate web pages efficiently and accurately.

Junjie Song, Jin Liu, Yuhui Zheng

Ontology Construction Based on Deep Learning

With the development of information technology, ontology is widely applied to different areas has become an important technology in knowledge presenting, knowledge acquirement and application. This paper proposes a method of multi-ontology construction based on deep learning, which is based on a great amount of non-structured text. We apply this method to an experiment regarding to the domain of shipping industry (including ship, harbor, shipping line and etc.). And the result shows that it is capable of constructing multi-ontology automatically and effectively.

Jianan Wang, Jin Liu, Lei Kong

Personal Attributes Extraction in Chinese Text Based on Distant-Supervision and LSTM

In this paper, we proposed a distant-supervision approach to solve the problem of insufficient training corpus for extracting attribute from the unstructured text, by using the wiki infobox information to tag the Wikipedia text to get the training corpus. We consider the extract attribute as the sequence annotation question and use the wiki personal text as the training corpus. The clp-2014 task4 is used as the test corpus to test. The experiment result show that this method can enhance the quality of the attribute extraction.

Wenxi Yao, Jin Liu, Zehuan Cai

Web Pages Ranking with Domain Ontology

With the growth of internet, web pages ranking algorithm is becoming increasingly important for search engines. The traditional web page ranking algorithm is based on hyperlinks without considering the semantic information of the web pages, which result in inaccurate search results. In recent years, researchers have tried various methods to understand the content of web pages, and have proposed many web page sorting algorithms based on semantic understanding. In this paper, a web page ranking algorithm called DOPR based on domain ontology is proposed. The results show that DOPR has higher accuracy and feasibility than other ranking algorithms.

Mingji Zhou, Jin Liu, Yuhui Zheng

Simulation-Based Reliability Improvement Factor for Safety-Critical Embedded Systems

In the design of safety-critical embedded systems (SCES), the use of reliability measures is crucial to identify reliability-optimized and cost-optimized fault-tolerant mechanisms (FTM). The reliability improvement factor (RIF) was used in this study, which is a ratio of the probability of failure of the baseline system to that of the redundant system for a fixed mission time. We extend the analytical RIF into the simulation-based RIF (SRIF), as a relative measure of the reliability improvement for the FTM of SCES. We calculated the SRIF of the FTM by substituting the failure rate, which can be obtained from the statistical fault injection simulation by using co-simulation models and representative fault models. We use SRIF to compare the performance of FTMs and find the most reliable FTM. As a case study, we compare the SRIF of the dual-modular redundant (DMR) FTM with the triple-modular redundant (TMR) using ARM7 SystemC simulation models.

Jongwhoa Na, Dongwoo Lee

SBraille: A New Braille Input Method for Mobile Devices

There are not many accessibility technologies to input braille in the mobile devices for the visually impaired. We have studied three conventional methods to input braille in mobile devices and, as a result of our study, we propose SBraille, a new braille input method using swiping gesture for mobile devices. In order to input a 6-dot Braille letter using SBraille, six dots are divided into three parts with two dots. Each part can be one of the patterns: ‘●○’, ‘○●’, ‘●●’, and ‘○○’. The first two patterns can be represented by tapping left side and right side, respectively. The last two patterns by swiping upward and swiping downward, respectively. The most important advantage of the SBraille method is that the visually impaired can input braille with only one hand, so that they can use the other hand for another use, e.g., holding the cane. Other advantages of SBraille and its comparative analysis with other braille input methods (Braille Keyboard, ThaiBraille, and MBraille) are also explained in the paper.

Soonyong Lee, Ji Su Park, Jin Gon Shon

Confirmation Delay Prediction of Transactions in the Bitcoin Network

Bitcoin is currently the most popular digital currency. It operates on a decentralised peer-to-peer network using an open source cryptographic protocol. In this work, we create a model of the selection process performed by mining pools on the set of unconfirmed transactions and then attempt to predict if an unconfirmed transaction will be part of the next block by treating it as a supervised classification problem. We identified a vector of features obtained through service monitoring of the Bitcoin transaction network and performed our experiments on a publicly available dataset of Bitcoin transaction.

Beltran Fiz, Stefan Hommes, Radu State

Failure Analysis in Safety Critical Systems Using Failure State Machine

In this paper, failure analysis of a railway level crossing system is studied using failure state machine. It was previously perceived that formal verification of safety critical system is possible using model checking and safety analysis technique [1]. Thus, in this study, we introduce some failure case study in previous approach [1] and failure analysis is accessed using the model checking counterexample. From the counterexample, we have proposed failure state machine for the failure analysis. From the findings, the need for design improvement is recommended.

Anit Thapaliya, Daehui Jeong, Gihwon Kwon

Simple and Low-Cost Heartbeat-Based Dual Modular Redundant Systems for Wireless Sensor Networks

Sensor node lifetimes have become a critical issue in the operation of wireless sensor networks (WSNs). The redundancy technique can improve the reliability and availability of WSN sensor nodes. However, augmenting sensor nodes with a redundancy mechanism increases the cost and performance of the system. In this paper, to improve sensor node reliability without increasing costs, we investigate three dual modular redundancy systems using low-cost hardware and a heartbeat mechanism with a UART and direct UDP channels. The proposed systems exhibit three advantages – low latency, cost-effectiveness, and easy configuration. We find that the average failover time for our DMR implementation using the direct UDP connection is 379 ms.

Jongwhoa Na, Dongwoo Lee, Munkh Zorigbold, Dongmin Lee, Sungyup Moon

Hidden Markov Model for Floating Car Trajectory Map Matching

Map matching is the key technology in the data processing of floating car trajectory data. In order to improve the matching accuracy, this paper adopted widely followed Hidden Markov Model (HMM) approach and proposed new probabilistic models for the transition probability. The new model considers distance difference feature and average speed difference feature, which was proved to be more reasonable and accurate to describe the context relationship between adjacent candidate points by experiments. The experiments showed that our proposed algorithm can achieve a better matching accuracy compared with a comparable HMM-based method from the literature.

Chengbo Song, Xuefeng Yan

Simulation Game System: A Possible Way to Realize Intelligent Command and Control

Following AlphaGo’s victory, ALPHA once again beat human pilots, bringing hope of intelligent command and control. In peacetime, however, sample data of military operational command and control is of limitation, causing problem for machine learning. Getting inspired from AlphaGo, Deep Green and ALPHA, a method is proposed to develop high fidelity simulation game systems, in order to accumulate sample data for machine learning, as a possible way to realize intelligent command and control, with some guiding significance to command and control technology development.

Xin Jin

Experiential Interaction Modeling for Virtual Training of Ultra-High Voltage Power System and Its Application

Staff skills training of Ultra-High voltage (UHV) power system have an urgent need for virtual reality (VR) simulation technique, and the interaction modeling is the critical factor to decide the effectiveness of training system. In this paper, the technical idea of user experience design based on flow theory is integrated to interaction modeling, which guiding the construction of the general technical framework of experiential interaction modeling. And three technical realization methods are proposed to improve user learning experience from different aspects, including building the optimal virtual environment, friendly interactive controlling and more attractive training mission design. The proposed modeling technique method has been practical applied in State Grid Jiangsu Electric Power Company, and the effective application results have verified the rationality and effectiveness of the proposed method.

Jian Shao, Wei Liu, Wei Dai, Shengkun Ma, Qun Li, Ji Wang, Jian Chen, Neal Xiong

Analysis on the Technology of the Internetware Comprehensive Testing

Based on the analysis of the basic characteristics of the network architecture, the technical problems faced by the internetware comprehensive testing are analyzed, and the testing requirements and testing steps of the internetware comprehensive testing are described from the aspects of requirements of it. Then, a specific method for internetware comprehensive testing are put forward from the perspectives of application components testing, application component interaction and integration testing, the system testing, which provides technical guidance for internetware comprehensive testing.

Zhengxian Wei, Mingqi Fan, Min Song, Hongbin Wang, Zhe Zhang

Analysis and Inspiration to Intelligent Command and Control

As the fast development of the artificial intelligence, military command and control system is trend to be more and more intelligent. The requirements of the intelligent command and control system are analyzed in this paper. The U.S. has been the front of the use of artificial intelligence in military. The Deep-green program, the simulated aerial warfare system ALPHA and the Commander’s Virtual Staff (CVS) program are analyzed both in technology and capability. Inspirations and suggestions were given to the development of intelligent command and control system.

Tingting Li

Exploratory Search for Learning: Finding the Concept with Minimal Cognitive Load

In this paper, an exploratory search selective query recommendation method is introduced to improve the efficiency of searching. A novel model based on user ability is proposed to minimize the user’s cognitive load, and the experiment results show that this method could achieve a good effect.

Zhuyin Xue, Zhen Hu, Yunhai Jia

Predicting the Metro Passengers Flow by Long-Short Term Memory

In this paper, LSTM is proposed to predict metro passengers flow to avoid traffic jams for the city governors. The model is validated by manual counted data and the results show that LSTM can report an instructive prediction.

Zhen Hu, Yi Zuo, Zhuyin Xue, Wenting Ma, Guilin Zhang

Research on Hierarchical Aggregation Method for Situation Assessment

The issue of the target aggregation is an important function which the situation assessment needed to implement. Because of such a variety of targets, complicated coordinated relationships and fast evolved battlefield situation in joint command and control, it is difficult for commanders to make effective decisions confronted with excessive information. In this paper, an analysis of target aggregation in situation understanding is made, and a mathematic model of the armored targets on the battlefield is built. On this basis, a hierarchical aggregation algorithm is proposed, and the information of operational units is classified in order to form the hypothesis of the military systematic units at relationship level, and to reveal the relationship between situation elements and functions of situation elements. Finally, the feasibility of the algorithm is verified through a situation example, thus laying the foundation for the intention of the target behavior judgment and the enemy combat attempt.

Xiaoxuan Wang, Zhenyi Zhao

The Testing Execution Mechanism on Internetware Oriented Flow Dynamic Building

Based on the DDS middleware with publish and subscribe mechanism, we propose a Flow Model of Internetware oriented dynamic building (FMoI). Through FMoI, the business process of the network software can be flexible organized, the criteria to judge whether the network software is running normally is established, and the correctness of process is built also. In order to support the automatic online testing of the network software, this paper puts forward an Execution Mechanism on Internetware Online Testing, called EMIOT, which describes the framework and process of implementing the online test execution mechanism of the network software on the DDS middleware.

Min Song, Ying Song, Zhengxian Wei, Hongbin Wang, Albert M. K. Cheng

A Model and Application of Collaborative Simulation Training System for Substation Based on Virtual Reality

Regarding the imperfection and poor collaborative interaction ability in existing substation simulation training system, this paper proposes a virtual reality simulation training system model for multiuser cooperation in substation maintenance which realizes multiuser collaborative virtual maintenance training of large-scale complex scenes of the substation and solves the competition of the electrical equipment resources. Through analysis of hybrid heterogeneous hierarchical process modeling technology and Petri Nets sharing sub-network synthesis technology, the key technologies such as hierarchical maintenance process modeling, distributed cooperative concurrency control and asynchronous message delay strategy was put forward. The proposed model technique has been practical applied in virtual substation training system, and the effective application results have verified the rationality and effectiveness of the model.

Wei Dai, Wei Liu, Xin-dong Zhao, Jian Shao, Qun Li, Sheng-kun Ma, Neal Xiong, Chang-nian Lin

A Model-Based Transformation from SCR Specification Models into Altatica3.0 Design Models

In the safety critical field, Model-based system safety analysis and verification has become an important methodology. In this paper, by using SCR specification of four variables model, and translate this specification into AltaRica3.0 model. We studied the relationship between the semantic of four-variable model and AltaRica3.0 model, and put forward transformation rules. Finally, we give a case study of a wheel brake system (WBS) to show the process of transformation and validation.

Jun Hu, Mingming Wang, Weijun Zhang, Wanqian Li

Simulation and Verification of Software Architecture for Loosely-Coupled Distributed System

Loosely-Coupled Distributed System has extensive application in data-intensive real-time fields; How to verify the function and performance of software architecture is an important and valuable issue. Aimed at this problem, a verification method is proposed in this paper. Firstly, the corresponding system model is established for describing the system. Secondly, according to the requirements of the verified system, the corresponding simulation components are automatically generated and the system is deployed; The simulation of the verified system is completed and the rationality of the system is evaluated and optimized. Finally, according to the development progress, the system gradually replaces the simulation components with real components and the final verification of the system is realized.

Shen Jun, Shen Xuan, Gao Wen, Zhang Yong

Research on Hybrid Storage Method of Massive Heterogeneous Data for Mobile Environment

The article relates to a hybrid storage system and method for processing massive heterogeneous data, mainly for the real-time information collection, high-speed storage and timely indexing in the mobile environment. The article provides a common heterogeneous data resource metadata description model, data mixed storage solution to standardize the sharing process of massive heterogeneous data resources, and provides an optimized index construction algorithm and data archiving method to realize the collection of massive data, index building and persistence, in order to complete the data resources sharing and effective use in mobile environment. The method in the article can effectively deal with various complex problems in mobile environment, for example, the heterogeneous data structure, a huge number, scattered physical location, the data complex content and so on.

Shanshan Wu, Fan Yi

User Targeting SaaS Application Interworking Service Framework Using Complex Context and Rule-Matrix

This paper presents complex contexts, rule-matrix and SaaS application inter-working service framework for user targeting service. They are key components in order to decide what SaaS service are needed and how those SaaS application are interworked for user’s convenience. The user targeting is performed according to user’s property and status of user. The complex context is set of information about who user is, what status is and what SaaS application is. The complex context consists of User Description (UD), Status Description (SD), SaaS Application Description (SSD). Simply, UD is what user is, SD is what user’s status is, and SSD is what property of SaaS Application is. The rule-matrix is a set of rules that explain how each SaaS application is interoperated. The SaaS application interworking service framework is an architecture that include all modules for managing SaaS application inter-working service. It collects complex contexts (UD, SD, SSD) in order to choose which SaaS applications are useful, analyze what functions to use, how to interwork and recommend the best rule to user.

Jong-Jin Jung, Yun Cui, Myungjin Kim

A Framework for Preventing Illegitimate e-Prescribing Practices

The e-prescribing system was introduced to improve convenience for patients when they use healthcare services. However, security is still a critical aspect because illegitimate e-prescribing systems can lead to serious problems. This paper aimed to propose a framework for preventing illegal issuance e-prescribing practices by applying case-by-case approaches and a cloud-based national prescription database.

Deuk-Hun Kim, Jin Kwak

Research on Horizontal Damage Zone Airspace of Ship-to-Air Missile Under Cooperative Operation

Horizontal damage zone is an important index to reflect the anti missile capability of ship-to-air missile. First, research the ship-to-air missile’s horizontal damage zone problem. By the ship-to-air missile and target as far as the meeting point determine the boundary of horizontal damage zone. Then, analyzing the characteristics of different cooperative mode operation, and according to the different mode of cooperation gives the method of calculation of ship-to-air missile’s horizontal damage zone. Last, giving an example of formation air defense operation, by Matlab simulation verifies the correctness of the conclusion.

Yitao Wang, Huayang Wang, Haiming Liang, Liying Wang

Key Issues of Military Simulation and Analysis System

Military complex system is a hotspot of current operational research. Due to the non-reproducibility of complex system and various complexity, traditional observation, experiment and other methods cannot effectively carry out the research, while simulation technology has become an important means of delving into complex systems. This paper describes our concept of military simulation and analysis system (MSAS), and how this system provides more accurate analysis and forecast methods to achieve better credibility of the results, aiding decide-making and reducing the decide risk for military complex system. The key issues of MSAS are discussed, especially system infrastructure, military model infrastructure and data. The paper then presents some observations and areas for further research.

Yitao Wang, Huayang Wang, Xinye Zhao, Kedi Huang, Haiming Liang

A Study on the Comparison of “Internet+” Paradigm and Means and the Analysis and Evaluation of Its Typical Development Path

Taking Internet as carrier and adopting emerging information and communication technology, intelligent science and technology, “Internet+” paradigm and means are emerging and applied to the paradigm innovation, business optimization and efficiency improvement. In this paper, comparative analysis is firstly conducted on the “Internet+” paradigm and means from the perspective of paradigm, means and support technology to propose that Smart cloud manufacturing (Smart CMfg) is a scheme to match Chinese national conditions and competitive advantages. From the perspective of model-based system engineering, the incentive and feedback model of manufacturing industry development is then proposed to discuss the development path and mechanism of the transformation and upgrading of Chinese Smart CMfg-based manufacturing industry. Finally, this article demonstrates that the operation of the incentive and feedback model of manufacturing industry development may form massive cloud enterprise network which could support the upgrading and development of Chinese manufacturing industry.

Chao Geng, Shiyou Qu, Guoqiang Shi, Baocun Hou, Mei Wang, Tingyu Lin, Yingying Xiao, Zhengxuan Jia

Stripe Noise Correction for Infrared Imaging System Using Neural Network Theory

For staring infrared image, there is inevitable stripe noise which seriously degrades the quality of the image. In this paper, a stripe noise correction is proposed using high frequency neural network theory. Firstly, the mechanism of production for stripe noise is discussed; secondly, the correction model of high frequency neural network is established, the gain and offset parameters are estimated; finally, stripe noise is corrected. Experimental results of real and simulated images show that the proposed method can correct stripe noise effectively, without producing ghosting artifacts.

Junqi Bai, Chuanwen Chang, Wen Liu

A Smart Indoor Gardening System Using IoT Technology

Regarding the pressing social concerns of well-being and aging society, this paper explores the potentials of a smart indoor gardening system that links gardening activities and the IoT technology. The benefits of gardening have been emphasized to improve the quality of life. However, a number of reasons due to a lack of gardening culture and apartment housing systems limit personal gardening in Korea. Thus, indoor gardening has been paid attention as an alternative, but it is still a challenge. Previous studies and practices have shown that the IoT technology can be applied to numerous occasions, and can feasibly provide a solution to these gardening issues. This paper proposes a possible smart indoor gardening system to cope with the issues.

Byoungwook Min, Sung Jun Park

A User-Defined Code Reinforcement Technology Based on LLVM-Obfuscator

With the popularity of embedded devices in daily life, the gap of hardware configuration is gradually narrowing, more and more differentiated functions are realized by software. How to effectively protect the intellectual property rights of software becomes very important. Software security issues of embedded software include reverse-engineering, malicious modifications and tampering. At present, most of the pure software protection solution is relatively simple, the protective effect of which is not desirable, while the hardware solutions have non-negligible costs. In this paper we discuss a user-defined code reinforcement technique based on LLVM-Obfuscator. Data and control flow transformation techniques are added based on the common code obfuscation techniques, and Hardware reinforcement solution is simulated by software. So that it does not increase the cost, but has better protective effect than ordinary software.

Xue Yao, Bin Li, Yahong Sun

An iPOJO Components-Based Workflow Architecture in Ubiquitous Cloud Environments

The growth of innovative services from various ubiquitous sensors and devices provides huge potential for ubiquitous computing applications and systems. However, to be able to benefit from the new developments, current service middleware technology needs to catch up with the rise. One of the effective solutions is to build up value-added, composite services out of basic services available in ubiquitous environments. Thus, this paper proposes to extend the iPOJO, a service-oriented component model, to enable a workflow engine, achieving service composition in OSGi-enabled ubiquitous environments. The features of modularity and service orientation have been the main emphasis of our workflow system design.

Xipu Zhang, Choonhwa Lee, Bleza Takouda, Ryu Giha

An Effective Framework for Identifying Good XML Feedback Fragments

In Pseudo relevance feedback, It is often crucial to identify those good feedback documents from which useful expansion terms can be added to the query. For Extensible Markup Language (XML) data, this paper proposes an approach for identifying good feedback fragments by a complete framework in which two phrases are included. (1) The first phase is about XML search results clustering. We performed a k-medoid clustering algorithm to XML fragments based on an extended latent semantic indexing model. (2) The second phase is a two-stage ranking. Cluster ranking is performed in the first stage to select relevant clusters on the basis of cluster labelling, which is determined by extracted fragment keywords based on a combination of weight and context; fragment ranking is performed during the second stage where multiple features are used to identify high quality fragments from the previously obtained relevant clusters. Experimental results on standard INEX test data show that the proposed approach achieves statistically significant improvements over a strong original query results mechanism, ensuring high quality fragments for feedback.

Minjuan Zhong, Beiji Zou, Lei Wang, Shumei Liao, Naixue Xiong

Minimum-Cost Consensus Models for Group Decision Making Under Intuitionistic Fuzzy Environment

This paper mainly focuses on the consensus problems under fuzzy environment, in which experts’ original opinions take the form of intuitionistic fuzzy numbers. Based on the objective of minimizing the total consensus cost, we develop a novel intuitionistic minimum-cost consensus model (MCCM) in order to evaluate the deviation between individual opinions and group opinion. The proposed model can not only yield optimal adjusted individual opinions and consensus opinion, but also can explore index of each expert’s risk-bearing attitude. Additionally, some intuitionistic consensus models under WA operator and OWA operators are presented. With the help of multi-objective programming theory, linear-programming-based approaches are put forward to solve these consensus models. Finally, a numerical example is implemented to demonstrate the accuracy and effectiveness of the proposed models.

Yuanyuan He, Chengshan Qian, Neal N. Xiong

Research on Ontology-Based Data Fusion

The paper proposes an ontology-based multi-sensor data fusion model framework for the wide application of multi-sensor data fusion, which uses ontology as the semantics model of data in the feature level data fusion to solve the heterogeneous problem of multi-source data. In the framework, an effective data processing algorithm is presented to preserve a reliable confidence level for data in a dynamic environment based on the requirements of data timeliness in real-time data fusion systems. Considering the uncertainty of fuzzy information, Transferable Belief Model (TBM) is used in the decision level of data fusion to achieve multi-source heterogeneous distributed data fusion. Finally, the effectiveness of the fusion framework and algorithm is verified via an example instance of onboard sensors data fusion.

Shun Wang, Da-zhou Kang, Yan-hui Li, Zhe Zhang, Zheng-xian Wei

An Effective High Resolution Rainfall Estimation Based on Spatiotemporal Modeling

High resolution rainfall estimation is one of the most significant input for numerous meteorological applications, such as agricultural irrigation, water power generation, and flood warning. However, rainfall estimation is a challenging task because it subjects to various sources of errors. In this paper, an effective high resolution rainfall estimation system is presented which employs a spatiotemporal model named RANLIST. The merits of this system are listed as follows: (1) RANLIST, which exploits both spatial structure of multiple radar reflectivity factors and time-series information of rain processes, is superior to other methods for rainfall estimation. (2) RANLIST is used for rainfall estimation with temporal resolution of six minutes, while this system can estimate rainfall every minute which will do more help for coping with emergencies such as flood. Experiments have been implemented over radar-covered areas of Quanzhou and Hangzhou of China in June and July, 2014. Results show that the presented rainfall estimation system can obtain good performance with spatial resolution of 1 km × 1 km, temporal resolution of six minutes or one minutes.

Qiuming Kuang, Xuebing Yang, Wensheng Zhang, Guoping Zhang, Naixue Xiong

A General and Effective Network Failure Ant Colony Algorithm Based on Network Fault Location Methods

With the development and evolution of network technology, the normal operation of network equipment to meet the most basic needs of the needs of business users. Meanwhile, the key is the normal operation of network services. Therefore, this paper proposes to quickly find a network failure ant colony algorithm, which uses equipment pheromone concentration determines the strength of the network devices failure probability, according to the pheromone concentration construction business failed path.

Ruan Ling, Liu Changhua, Wang Yuling

Study on Smart Automated Sales System with Blockchain-Based Data Storage and Management

Automated sales systems, a recent technology, offer a wide range of products and services. This technology has advanced by combining with Internet of Things (IoT) technology. However, because of the development of such technology, data storage and management systems with trusted third parties cannot efficiently defend against data forgery attacks. In this study, a blockchain network is applied to an automated sales system. A smart automated sales machine using smart contracts is proposed. This system enables users to know the quantity of products or status of service provision before they visit an automated sales system. In addition, the system administrator can immediately see the expense-reduction effect of system management and address any problems in the system.

Minjae Yoo, Yoojae Won

Pairwise Relation Analysis and Quality Estimation of Classical Chinese Poetry in Ancient Korea

This paper represents a text mining-based analysis on Classical Chinese poetry in ancient Korea. We try to evaluate the relations between poems with several document similarity analysis methods, and also to estimate the quality of poems based on simple hypotheses. Nine poem books in the 15th century have been selected for analysis to validate the effectiveness of this approach. In order to overcome the limited number of data, we slightly change the existing similarity measure and try to adjust target data by considering the structure of Classical Chinese poetry. Analysis results show a high potential of this approach by producing outcomes that fairly match expectations.

Shohrukh Bekmirzaev, Byoung-Chan Lee, Tae-Hyong Kim

A Parallel Adaptive Circumference Method with OpenMP

The implementation of identifying regional and local gravity fields computing algorithm by adaptive circumference method is achieved in this paper, aiming to the drawback of traditional circumference method in a chosen suitable radius among varies objects. We get suitable radius automatically by the minimum variance of computing point relative to the points on circle in series of set radius, and use the average of regional field value instead of the average of anomaly gravity when calculated point is as the center and to get closer to actual fields in iteration procession. Further, a parallel algorithm based on OpenMP employed multi-core CPU has been presented for high efficiency. Otherwise, the relationship between the number of cores and parallel efficiency is analyzed. The results of theoretical model and actual data show not only our adaptive circumference method can clearly separate abnormal boundary and get accurate abnormity, but also the algorithm is adaptive and has high computational efficiency.

Hongtao Bai, Jinwei Fang, Fengxu Zhang, Xiaomeng Sun

Maximum Stack Memory Monitoring Method Assisted by Static Analysis of the Stack Usage Profile

As IoT permeates through industry in general, the safety assurances of IoT will become a major issue. One of the major safety issues, stack overflow, is a bothersome and difficult problem because it is hard to discover during design and to prevent. Many related studies for preventing stack overflow have used two general methods. The static analysis method is employed before a program runs and estimates the program’s probable maximum stack memory usage. The dynamic analysis method is used to monitor for stack overflows during run-time. Based on those prior works, this paper introduces a method for monitoring stack memory based on static analysis of the maximum stack memory usage profile. We anticipate that applying the proposed approach will prevent stack overflow in an efficient manner.

Kiho Choi, Seongseop Kim, Moon Gi Seok, Jeonghun Cho, Daejin Park

Design of Nonlinear Data-Based Wellness Content Recommendation Algorithm

As IT technology has advanced and people’s interest in wellness has increased, recommendation algorithms are being developed to allow people to use wellness content easily. However, existing recommendation algorithms use data entered by users and content-based filtering to recommend content, making it difficult to recommend areas of interest which change in real time. Therefore, in this paper we propose an algorithm which creates user information based on nonlinear social network data and makes recommendations in real time in order to reflect the user’s recent interests. The test result verified that the proposed algorithm improved accuracy by 31% compared to that of the existing content-based recommendation algorithm.

Young-Hwan Jang, Seung-Su Yang, Hyung-Joon Kim, Seok-Cheon Park

User Modeling Based on Smart Media Eye Tracking Depending on the Type of Interior Space

This study tried to combine user interior location positioning data using Wi-Fi RSS technology with remote type eye tracking data, analyze it, and propose user modeling depending on individual user inclination. It grasped user inclination depending on the distance between a user and smart media and the difference in the screen size by conducting an experiment on smartphone, tablet PC, and notebook using a single camera. In addition, it classified public places such as a cafe or a library into open space and individual places such as a personal learning room into closed space, and then derived their correlations with concentration considering the type of space and the number of persons existing in the space. Through a combination between location positioning data and eye tracking data, it could reflect user inclination well and prove that users would show different user behaviors depending on changes in media and place.

Hyejin Song, Nammee Moon

Frame Rate Control Buffer Management Technique for High-Quality Real-Time Video Conferencing System

The limitation of a real-time video conferencing system is that it does not perfectly guarantee real-time transmission due to a delay in the network and buffering as well as ineffective communication of user information between systems. Studies are actively investigating the network infrastructure expansion and jitter delay in order to overcome this problem. However, there has not been much progress with respect to buffering delay. This paper suggests a Frame Rate Control Buffer (FRCB) management technique to solve problems that occur due to buffering delay. The FRCB is used to prevent buffer overflow and underflow by adopting two levels of buffer thresholds, Fast-play Threshold (FTH) and Slow-play Threshold (STH). It demonstrates superior performance compared to jitter buffer in conditions such as high CPU load, thus proving its suitability for high-quality real-time video conferencing.

SangHyong Kim, Yoojae Won

Method to Modify the Hex of Android Manifest File in Android Apps for Dynamic Analysis

Analysts sometimes need to analyze the app depending on the situation. There are two main ways to analyze Android apps. This is the static analysis that grasps the flow of the app through the source code and the dynamic analysis that analyzes the variable that changes during the app’s operation. For dynamic analysis, this can be done by setting the debugging option of the Android Manifest file. In most cases, modification is done by decompiling the app and modifying the original source. In some cases, however, there is a problem with the decompiling process. So we propose a way to modify the debugging option of the Android manifest file to “true” without decompiling the app.

Suhyoo Lee, Junhoo Park, Jaecheol Ryou

An Implementation of Unidirectional Security Gateway to Guarantees Inter-VTS Exchange Service Security

The IVEF service developed by IALA provides a common framework for the exchange of vessel traffic information. But, it does not have a secure transmission function. When one VTS center is hacked, other VTS centers connected to IVEF are also threatened with hacking. In order to cope with security threats targeting the VTS system, there is a growing trend to use unidirectional security gateway. In this paper, we propose to prevent a threat to hacking each VTS center by exchanging IVEF information using the unidirectional security gateway.

Yong-Kyun Kim, Seoung-Hyeon Lee

Implementation of NFC-Based Smart-Drug Information Management System

This paper is a research paper that develops NFC-based smart-drug information management cap that can be easily attached to existing medication so that it can be easily used to provide information for elderly patients. We aim to develop a system prototype that provides smart-drug information management service so that elderly patients can easily receive drug time without alarm regardless of place and can check the drug history information at any time.

Kyeong-Rae Cho, Tae-Bok Yoon

DSS-SL: Dynamic Signage System Based on SDN with LiFi Communication for Smart Buildings

In this world, everyone is connected, and things become interactive. The eventual fate of the intuitive world relies upon the future Internet of Things (IoT). Smart building is an important aspect of the IoT. During traffic and evacuation, to facilitate the passage of occupants, signage systems are widely used in the built environment. Due to the complex layout of large buildings, rapid evacuation is making way finding difficult. The need for exit signs that take into account when this should be evident, redirecting individuals to not only an exit but rather a practical and optimal way out in an evolving emergency situation has facilitated the improvement of another era of cutting edge signage. Conventional signage cannot adapt to changing conditions because it only conveys single and passive information, and displays static content with limited user interactivity. To address this issue, in this paper, we propose a dynamic active emergency signage system architecture ‘DSS-SL’ based of SDN. We also present a low power LED-based visible lighting communication between smart devices and forwarding switches inside the building.

Pradip Kumar Sharma, Byoung Wook Kwon, Jong Hyuk Park

A Study of Service Quality in Multi Cloud Computing

Recently, the cloud services have largely increased due to smart working which allows people to work freely anywhere. For this reason, the volume and the types of data have also increased so that it has become an urgent priority to meet the functional requirements for the services. The solution can be cloud service. However, even the cloud services require a method to support the HW resources (e.g., memory, server or network) when the system experiences capacity deficits because of its limited capacity. Thus, so called a ‘multi-clouds’, which allows provision of needed resources by configuring a multiple number of clouds, is necessary. In this study, we present a method of measuring quality and the standard of service level when a multi-cloud service is required due to the limitations of a single-cloud which cannot deal with both capacity and service requirements. By supporting the service with multi-cloud resources that cannot be included in the single-cloud, a better quality service will be provided to the users. When a customer uses a single cloud, the service provider will not be able to provide unlimited service due to the limitation of available resources. The multi-cloud system can deal with this problem but a suitable quality assurance method should be secured as well. In the study, we have conducted a research on the method of securing improved SLA for multi-clouds.

Sangdo Lee, Yongtae Shin

Research on Cloud-Based on Web Application Malware Detection Methods

Recently, the use of the Internet has increased, and the spread of malicious code through web application vulnerabilities has become a major threat. In this paper, we propose a technology to protect web applications securely by detecting unknown malicious code through signature analysis of cloud based web application.

Ki-Hwan Kim, Dong-IL Lee, Yong-Tae Shin

Study on Customer Rating Using RFM and K-Means

The RFM (Recency, Frequency, Monetary) market analysis technique is a widely used in the marketing field to analyze customer behavior. The interest in machine learning has recently increased to utilize the increase in accumulated data. Therefore, an attempt was made to analyze data by combining the RFM technique and various algorithms. In this study, we attempted to classify customers through the RFM technique and k-means algorithm, which is a typical clustering algorithm. In a conventional experiment, there are many cases where the k value is designated as 8 or 9. However, in this experiment, the optimal k value for the data set was obtained using an internal evaluation method.

Hyunjung Ji, Gyeongil Shin, Dongil Shin, Dongkyoo Shin

Scalable Distributed Temporal Reasoning

In this paper, we propose the design and implementation of a large-scale qualitative temporal reasoner, MRQUTER, which can perform reasoning over large Web-scale knowledge bases. This temporal reasoner is built on a Hadoop cluster system using the MapReduce parallel programming framework. It decomposes the entire qualitative temporal reasoning process into several MapReduce jobs and incorporates some optimization techniques into each reasoning job component, implemented using a pair of Map and Reduce functions. Through experiments using large benchmarking temporal knowledge bases, MRQUTER shows high reasoning performance and scalability.

Jonghoon Kim, Incheol Kim

An Efficient Algorithm for Influence Maximization Based on Propagation Path Analysis

The problem of influence maximization is to find a subset of nodes in a social network which can make the influence spreading maximized. Although traditional centrality measures can better identify the influential nodes, there are still some disadvantages and limitations. In this paper, we firstly propose a propagation path model which can find m paths with the highest probability from a certain node to other nodes in the network. Then utilizing the propagation model, the node set that are most likely to activate a certain node can be obtained. By implementing simulations in three real networks, we verify that our proposed algorithm can outperform well-known centrality measures. We also use the independent cascade model (IC) to evaluate the spreading ability of nodes with different centrality measures. In comparison with traditional centrality methods, our method is more stable and generally applicable. Besides, we apply PPA method extended to signed networks where we proposed PPAS method, through the experiments on real data sets, our PPAS method has the better performance in identifying the state of the nodes in networks.

Wei Liu, Xin Chen, Bolun Chen, Jin Wang, Ling Chen

Data Stream Clustering Algorithm Based on Bucket Density for Intrusion Detection

The ability to process data streams has become one of the challenges of the current intrusion detection systems. A data stream clustering algorithm based on bucket density is proposed for this situation which is able to identify clusters in any shapes and the speed of online layer is fast. Feedback principle is used to solve the problem that some of the edge of the bucket is lost and users does not need to specify the number of clusters. An intrusion detection system is constructed with the improved algorithm. The experiment shows that the algorithm proposed has fast speed for clustering. The system based on the algorithm has a better capability of detection.

Chunyong Yin, Lian Xia, Jin Wang

A Study on Subsequence Similarity Join in Time Series Data Using MapReduce

There are a large number of applications that find the most similar pairs of time sequences in a given time-series database. However, similarity join operation in vast amounts of data is a big challenge in a single machine. For such data-intensive computing, distributed parallel processing framework such as MapReduce is getting a lot of attention. In this paper, we investigate how to operate subsequence similarity joins using MapReduce framework. We first show a sequential subsequence similarity join algorithm. Next, we propose two efficient algorithms to minimize the subsequence similarity join computation. We finally perform the experiments with synthetic data sets. The performance shows that the effectiveness of our MapReduce algorithms.

Kyounghyun Park, Hee Sun Won, Keun Ho Ryu

A Practical Study on Data Logger for Gas Industry

Industrial fields have been adopted wireless communication for efficient safety management. It is very dangerous that industrial fields try to apply generous wireless method without reliability and safety requirement. Especially, industrial gas safety fields are necessary more reliability and safety to wireless communication in order to prevent unexpected risk. This paper aims to develop practical use technology of remote wireless sensing system for gas facilities. At first, we evolve network management and connection for gas facilities in order to find, control and manage gas devices. Next, our study designs high reliability network adaptor with heterogeneous gas devices like as data logger as a sensor node. This data logger is divided into A type and B type with its intended use. Finally, we construct two test-beds for validating our results.

Jeong Seok Oh

Operational Reliability Analysis of Guided Weapon Systems

Reliability is the priority matter in one-shot guided weapon systems. The reliability prediction data is used during the development stage as the manufacturing cost is very high and the production quantity if quite limited. At the same time it takes relatively a long period of time to acquire a reliable operation data set after deployment such that in order to determine the operational reliability, weapons must be tested and analyzed in real operating environments. For the research, the life distributions were estimated by using actual operation data and the reliability was calculated by applying the method of least squares and maximum likelihood estimation. As a result, the actual reliability of each system was higher than predicted reliability. It was possible to confirm the actual operational reliability of domestic (ROK) guided weapon systems through this research and the methods used here will contribute to the reliability analyses for the future guided weapon systems to be developed.

Ju-seok Ha, Kyung-mo Kim

The Direction of Information Security Control Analysis Using Artificial Intelligence

The areas where artificial intelligence (AI) is employed are gradually increasing. The latest malicious codes are continually being found in the security control area and the security teams in various organizations are investigating an average of 200,000 security incidents a day and often wasting much of their time in tracking wrong targets or attacking methods. It is expected that the security-related incidents will be increased more than twice in near future. Thus, the security control staff will be able to prevent security breaches only by rapidly analyzing the latest vulnerabilities and logs in their systems or security equipment. In this study, we have studied the possibility of utilizing current AIs used for diagnosis of cancers, translations or simple conversations, along with the future direction of AI for security control. The study also attempts to find an effective method of reducing damages by rapidly analyzing attack methods and vulnerabilities, hoping the method will be effective in protecting the systems from a new variety of attacks.

Sangdo Lee, Yongtae Shin

Anonymous Signature with Signer-Controlled Opening Capability

Anonymous signature is a method that provides validities of signature while signers are not exposed. Group signature is a method in anonymous signatures that the verifiers can assure the signer is one of the members from a group while the specific signer is unknown. In the group signature, the identification of the anonymous signer can be revealed by openers. The fact that the openers can reach to the signer’s identifications from an anonymous signature will be a burden of the signer. This paper suggests the signer acquires a token when creating an anonymous signature and the token will allow the opener to check the signer’s identification.

Sungwook Eom, Jun-Ho Huh

Group Signature with Signer-Controlled Opening Capability: Separate Token Generator

One of the most important cytological authorization methods to protect privacy is group signature method. Verifiers have limited information of the signers that gives the limited certainties that the signer is from the group while the specific identities of the signers are still concealed. In a group signature method, the openers can reveal the anonymous signers. However, the point that opener is always authorizable to check the signer’s identification without consent is useful to manage malice actions but is also doubtful for the signers because of the risk of the privacy exposure. This paper proposes a resolution for the signer’s anxiety. It tells that the signer has a right to create a token whenever they want to open their identity and only the opener with the token from the signer have access to the signer’s identification.

Sungwook Eom, Jun-Ho Huh

The Personal Information Overloads Effect Information Protective Responses in the Internet of Thing (IoT) Era

With the emergence of information overload in the age of the Internet of Things (IoT), data are collected and information is processed regardless of personal will. This study aims to examine the impact of personal information overload on perceived risks and information protective responses in the IoT environment. In this study, a research model related to personal information overload was proposed and empirical analyses were conducted. The major findings are as follows. First, personal information overload had a significant effect on perceived risks (economic, social, time loss, and privacy risks). Second, among perceived risks, economic, social, and privacy risks had a significant effect on information protective responses while time loss risk did not.

Won-Hyun So, Ha-Kyun Kim

Development of Virtual Workbench to Solve Computational Science Problem

Workbench application service for computational science and engineering is usually constructed working in personal computer with a big data input query and output data for example mesh data. Likewise, web-based workbench simulation system (web based workbench application) is required because workbench simulation system has user-friendly UI (User Interface). In this paper, we designed and developed by each requirements and the kind of execution when user whoever access to workbench service on the web invoke simulation project or test their science application to solve a computational science application.

Yejin Kwon, Inho Jeon, Sik Lee, Kumwon Cho, Jerry H. Seo

A Design and Implementation of the CoAP Adaptor for Communication Between DDS-Based Adaptors and External Devices

With the growth of the Internet market, a number of communication protocols and corresponding equipment have developed. Due to heterogeneity issues with the use of various communication protocols, the interoperability between different devices, middleware or gateway based interoperable systems have been studied. In this paper we suggest a DDS (Data Distribution Service) based CoAP (Constrained Application Protocol) adaptor in order to solve the problem of the DDS middleware based interoperable system when used with external accessibility or resource-constrained devices. By using the CoAP adaptor, we provide a data transfer service where in previous researches were to be seen as hard to access.

Gil-Tak Oh, Moon-Ki Back, Kyu-Chul Lee

Data Set Construction and Performance Comparison of Machine Learning Algorithm for Detection of Unauthorized AP

With the frequent use of Wi-Fi and hotspots that provide a wireless Internet environment, awareness and threats to wireless AP security are steadily increasing. Especially when using unauthorized APs in company, government and military facilities, there is a high possibility of being subjected to various viruses and hacking attacks. Therefore, it is necessary to detect and detect authorized APs and unauthorized APs. In this paper, to detect authorized APs and unauthorized APs, the characteristics of RTT (Round Trip Time) values are set as dataset and each machine learning algorithm SVM (Support Vector Machine), J48 (C4.5), KNN (K nearest neighbors), and MLP (Multilayer Perceptron).

Doyeon Kim, Dongkyoo Shin, Dongil Shin

Distributed Approach for the Security of P2P Wireless Network

Security for mobile P2P networks represents an open research and a main challenge regarding to their vulnerability and convenience to different security attacks such as Sybil attacks, black holes, etc. In this paper, we propose a solution based on a modification of the AODV routing protocol, taking into account the behavior of each node participating in the network solution. The benefits of our proposal are evaluated by simulation.

Chunyong Yin, Nimenya Stacey, Tatiana Moreira Beita, Jin Wang

Build Chinese Language Model with Recurrent Neural Network

In recent years, the introduction of Deep Learning based machine learning methods have greatly enhanced the performance of Natural Language Processing (NLP). However, most Deep Learning based NLP studies in the literature are aimed at the Latin family languages. There is seldom research which takes Chinese language model as the objective. In this paper, we use Deep Learning method to build language model for Chinese. In our model, the Fully-connected Neural Network which is a popular structure used in NLP is replaced by the Recurrent Neural Network to build a better language model. In the experiments, we compare and summarize the differences between the results obtained by using the original Deep Learning method and our model. And the results prove the effectiveness of our proposed model.

Li Lin, Jin Liu, Zhenkai Gu, Zelun Zhang, Haoliang Ren

Pedestrian-Safe Smart Crossing System Based on IoT with Object Tracking

Crosswalk is a way to sharing the road between vehicle and people. As the population increases, the chance of accident rises. Due to efforts on decreasing accidents, the total number of accident decreases annually. Nevertheless, the number of pedestrian accident in crosswalks is not decreased for a decade. In this paper, we propose a Pedestrian-Safe Smart Crosswalk System based on IoT using a CCTV with object tracking to reduce and save lives from an accident.

KwangEun An, Sung Won Lee, Young Ju Jeong, Dongmahn Seo

Management and Measurement of Firm Smart Business Capability

The smart business capability of a firm is very critical for the efficient execution of its business activities and for effective improvement of the performance of business tasks in a global business environment. In this business environment, the management and measurement of a firm smart business capability need to efficiently establish and improve the smart business environment appropriate for its management strategy and major business departments. This research presents a comprehensive measure of a firm smart business capability to totally manage and improve its smart business capability. The validity and reliability of the developed 14-item scale is verified by factor analysis and reliability analysis based on previous literature.

Chui Young Yoon

Study on Malicious Code Behavior Detection Using Windows Filter Driver and API Call Sequence

As the internet environment has been developed recently, threats and damage to malicious codes are increasing day by day. Most of the damage is caused by new and variant malicious codes because of the vulnerability of Endpoint. Most of the Anti-Virus used in endpoints run on a signature basis, and as intelligence on malicious code is developed, the detection rate of existing Anti-Virus is declining. Therefore, there is a need for a technology capable of handling new and variant malicious codes in real time on the endpoint. In this paper, we present a method for analyzing behaviors of malicious code using behavioral analysis of the Windows kernel function call sequence.

Kangsik Shin, Yoojae Won

A Study on Optimal Warranty Period for Repairable Weapon Systems

To the manufacturers, the warranty period is an important decision to make as it involves warranty costs. Although it can be decided by considering marketing strategy in general consumer product industry, it is relatively difficult to decide an adequate period of warranty for a G2B transaction. In this study, we have analyzed the life (failure–time) distribution for each item traded in G2B transactions. Also, their warranty costs based on warranty terms were calculated by applying the Warranty-Cost models. Finally, an optimal warranty period was determined for each item by applying a sale–A/S expenditure limit ratio. As a result, the necessity of setting different warranty policies and periods has been confirmed and established a basis for providing a reasonable warranty terms in the future for similar products.

Ju-seok Ha, Kyung-mo Kim

A Development on the SaaS Cloud Scientific Computing Platform for Education and Research in Nanoscience

We have designed a Software as a Service (SaaS), cloud scientific computing platform for education and research in nanoscience areas. Being named as EDISON (EDucation-research Integration through Simulation On the Net) Nanophysics, the service framework presents various Technology Computed-Aided Design (TCAD) software, with which users can study fundamentals of physics, explore properties of nanostructure materials, and design semiconductor devices through using free and easy access web-browsers. Here, we discuss technical focuses of the EDISON Nanophysics with an introduction of its service-status, upon the opening of a service in September 2012.

Inho Jeon, Jin Ma, Jerry H. Seo, Jongsuk Ruth Lee, Kumwon Cho, Hoon Ryu

EEG Based Smart Driving for Intelligent Accident Management

Electroencephalography (EEG) is one way of conveying thoughts and emotions and can be usefully analyzed and acquired. It is also emerging as a means of connecting people and things after speech recognition technology. EEG technology is used in a variety of fields, mainly in health and medical applications. Currently, it is converged and applied to various industries such as information and communication equipment, aircraft, and clothing items according to convenience of people. In particular, this technology is used as a technique for safe driving of drivers, and if commercialized, life can be protected even in emergency situations. In this paper, we can measure the driver ‘s brain wave and then find out the driver’ s condition based on the EEG data. Also, in case of emergency, it is possible to respond quickly and effectively.

Byung Wook Kwon, Jong Hyuk Park

DOTP-AaaS: Dynamic One Time Password Matching-Based Authentication as a Service

Recently, various devices have appeared through rapid development, and various damages such as leakage of personal information and hacking have occurred. So as the need for authentication security emerges, integrity is demonstrated using Secure Communication Dynamic One Time Password (DOTP) Matching. Based on this, it is expected to be applied to Authentication as a Service, which will have advantages of user convenience, simplification of authentication method, reduction of introduction cost and maintenance cost, and usefulness of management. In this paper, we have studied Authentication as a Service for matching secure OTP tokens based on Dynamic OTP Matching. It is expected to be used for SNS and Bank through Authentication as a Service which includes various services.

Nam Yong Kim, Kyung Yeob Park, Jong Hyuk Park

The Study on Data of Smart Home System as Digital Evidence

The Internet of Things (IoT) presents many possibilities, including security and privacy issues. The Digital Forensics has long been studied in academia and industry, but forensics for smart home device has never been attempted. Smart home forensics deals with tools and techniques for recovering data and evidence from mobile devices. This paper describes a data acquisition, classification and analysis process of smart home devices using the IoT. It also includes analysis based on attack scenarios of collected data and smart home device forensic models suitable for such scenarios.

Jung Hyun Ryu, Seo Yeon Moon, Jong Hyuk Park

A Solution for Reducing Redistribution Costs of HAIL

The Proof of Retrievability (PoR) is a useful tool for securing data by monitoring the retrievability of a file stored in remote servers. But they are not secure if the full data stored in the storage server is attacked. HAIL has been proposed to solve this problem. It enables a client to verify that files stored in independent storage servers are intact and retrievable. If some servers are attacked, a client can reconstruct the original data using the data stored in remaining servers. Unfortunately, in HAIL, expensive redistribution costs occur if we need to reconstruct the original data, which was not considered in existing works. In this paper, we propose a solution for reducing redistribution costs by grouping file segments and applying HAIL to each group. It is unnecessary for clients to download all files in all servers when some files stored in a specific server are corrupted. Finally, we analyze the performance of our scheme.

Taehyuk Kim, Minseok Lee, Doo Ho Choi, Taek-Young Youn

New User Management Technique in Storage Services for Stronger Privacy

Until now, online services including storage services have revealed their clients’ private information in various ways. Regardless of the reason of exposure, the damage caused by such accident is very serious. To support user-specific services, service providers maintain user-related information and the information can be harmful for the privacy of clients when the information is revealed to an adversary. To get rid of the source of the problem, in this work, we will introduce a new technique which permits service providers to support their clients without maintaining user-related information. The basic idea of our work to support the service provider has a set of parameters which are used for verifying the proof generated by a client as an evidence of his right regarding a service. To embody the idea, we will use a group signature scheme which is widely used as a tool for verifying someone’s membership instead of his specific identity. For the security of our scheme, we give provable security of our scheme under formal security models.

Taek-Young Youn, Ku-Young Chang

Gesture-Based User Interface Design for UAV Controls

Recently, unmanned aerial vehicles (UAVs) are being used to prevent or cope with disasters such as fire. When a disaster occurs, the photographs of the disaster site can be captured with an UAV to identify the disaster site. However, in order for UAVs to perform such roles as delivery of relief goods to the disaster site as well as taking photographs of the site simultaneously, multi-functional control methods are required. This paper proposes a method of controlling multiple functions provided by the UAV according to the user’s gestures through a gesture-based user interface. In the experiment, the control signals of UAV were generated by recognizing the user’s gestures using the HTC VIVE, which is one of haptic devices.

Jeonghoon Kwak, Yunsick Sung

Communication System of e-Navigation Between Vessel and Shore Utilizing Representational State Transfer at Sea

Currently, maritime wireless communications have many limiting factors. In particular, a damping effect of wireless communication, a problem caused by fixed installations, and a distance from other nearby communication. Among the existing wireless communication, satellite communication is the most common communication method in terms of convenience compared to the distance. However, considering the domestic economic and communication environment conditions, it requires high cost to use. Therefore, the Ministry of Maritime Affairs and Fisheries is developing efficient communication at sea in case of disaster by using Long Term Evaluation-Maritime (LTE-M), But in communication using only LTE-M, it is difficult to support perfectly in a disaster situation, and there are many difficulties in connection with communication such as Automation Identification System (AIS). In this paper, we propose a system that can support LTE-M without imposing much burden on existing system by configuring local network using LTE-M in the maritime disaster situation and expect contributing to the maritime wireless communication environment using the AIS system in LTE-M environment efficiently.

Teahoon Koh, Yonghoon Kim, Kamyoung Park, Jeongho Lee, Kyungryong Seo

Continuous-Time Estimation Filtering with Incorporation of Temporary Model Uncertainty

In this paper, a continuous-time estimation filtering is developed to incorporate temporary model uncertainty. The infinite memory structure (IMS) estimation filter is applied for the certain system and the finite memory structure (FMS) estimation filter is applied for the temporarily uncertain system, selectively. Therefore, one of two filtered estimates is selected as the valid estimate according to presence or absence of uncertainty. In order to indicate presence or absence of uncertainty and select the valid filtered estimate from IMS and FMS filtered estimates, two test variables and detection rule are defined. Computer simulations show that the proposed continuous-time estimation filter works well for both certain system and temporarily uncertain system.

Pyung Soo Kim

Adverstise Based Adaptive Model for IoT Device in Network Virtualization Environment

Internet of Things (IoT) systems are inherently dynamic. The number of such systems has been grow so fast. Network virtualization technology has changed the networking industry dramatically. In this paper, we introduce a configuration of common device metadata (CDM) for IoT device including device sensing data configuration, device communication and non-functional requirement. We discuss an advertise-based adaptive model for IoT devices composed of soft-sensor for the implementation of services like data source monitoring.

YunHee Kang, Younhoon Park, Jonghee Yoon, KwngMan Ko

Design and Implementation of a Wearable Device for the Blind by Using Deep Learning Based Object Recognition

Recently, deep learning based object recognition systems are very widely used in various fields, including surveillance systems. The accuracy of object recognition based on deep learning is better than other schemes. In this paper, we propose a wearable device for the blind by using deep learning based object recognition. Based on the implemented prototype and evaluation results, we confirmed the usefulness and effectiveness of the proposed wearable device.

Bongjae Kim, Hyeontae Seo, Jeong-Dong Kim

A Preference Based Recommendation System Design Through Eye-Tracking and Social Behavior Analysis

The study of recommendation services based on eye-tracking and social behavior analysis was conducted using either implicit or explicit data, and thus, carried the disadvantage of a decreased recommendation accuracy, having failed to supplement the flaws of each type of data. Therefore, the present study proposes a system applicable to recommendation services after deducting the personal preferences of the user by combining and analyzing the implicit data of eye-tracking and personal social behavior data with the explicit data of purchase data. By conducting experiments capable of obtaining category preferences based on smart phones, tablet PC, and smart TV, the study confirms changing preferences following the characteristics of the smart device. Ultimately, the study attempts to increase the accuracy of recommendations by using both implicit and explicit data and to achieve a recommendation system based on a collaborative filtering that considers device characteristics.

Heyjin Song, Nammee Moon

Design of Consumer Behavior Analysis by Region Through Reflecting Social Atmosphere Based on SNS

Consumption analysis research has been so far carried out only with existing statistics and data, and it has been researched without considering real time issues. Therefore, in this study we present an analytical method which reflects both non-real-time and real-time consumption behavior by region. Consumption behavior by region is extracted with sequential pattern mining based on PrifixSpan by combining location and consumption data based on time. Also, non-real-time data (Card consumption statistics) and real-time data (SNS Data) are analyzed by examining consumption ratios of six consumption category by region. Finally, the analysis is performed by calculating the consumption figures by each region of non-real-time and real-time data in accordance with the consumption behavior ratios extracted by region. This method is meaningful as it does not only reflect regional consumption characteristics, but also reflect both non-real-time and real-time, and it is expected that we can utilize when we research various recommendation services in the future.

Jinah Kim, Nammee Moon

Glove Type Air Mouse Powered by Kalman Filtering and Complementary Filtering

Recently, an air mouse has been used for preventing wrist syndrome or for convenience of use. However, in the case of the conventional air mouse, most of it operates as a remote control type. In this study, a glove type air mouse was developed. We used Kalman filtering and complementary filtering in order to perform mouse operation properly. The Kalman filtering improves the accuracy of the measured data values, and the complementary filtering accurately transforms the 3D data into the 2D data. And, the glove-shaped air mouse is configured to include the function of the pen and the function of the laser point. This shows that it is more efficient than previous remote mouse.

Jae Sung Choi, Won Jun Byoun, Ji Su Park, Min Hyung Lee, Ye Seul Kang, Hyun Lee

An Indoor Positioning System Using RSSI and BSSID

Research on indoor positioning technology has been carried out using various methods including WiFi, BLE, and terrestrial magnetism. The present proposes a system providing location information by collecting WiFi data (RSSI, BSSID). The proposed system applies an ensemble learning method of RandomForest to compare differentiation performance. RSSI and BSSID were used as differentiation performance variables, and the proposed indoor space was divided into grid shapes for the experiment. The algorithm presented in the experiment proves improved speed and accuracy compared to a RandomForest method that uses RSSI alone. The present study is expected to be utilized in the fields of indoor navigation and emergency rescue.

Sunmin Lee, Nammee Moon

IoT-Based VR Service Model to Improve Exercise Capacity

Recently, Internet of Thing has been actively studied in various fields such as wearable devices, smart cars, and smart factories. According to the Gartner report of 2017, the IoT market is expected to grow by more than $ 2 trillion by 2020. In this paper, we aim to develop a VR service model that combines healthcare and virtual reality through IoT technology to improve exercise capacity. In other words, we propose VR service model that can improve the exercise needs of the user by using bicycle which can be easily accessed by people for improving the IoT based athletic performance and incorporating VR service. The proposed model improves the sense of reality through 3D modeling using virtual reality technology, and the user can intuitively confirm the driving record and driving information, and can increase the efficiency of the user’s motion through the target heart rate.

Jeong-Dong Kim, Min-Gyu Park, Do-Yeon Ki, Bum-Hee Cho, Gil-Yong Lee, Bongjae Kim

Traffic Prediction System Utilizing Application and Control of Environmental Information

In the inference analysis using structured data, it can be thought that the analysis is easy because the meaning of the formal data can be accurately determined. However, it can be very difficult to predict the area whose data is sensitive to the surrounding environment such as traffic volume. To solve this problem, we proposed a system which improved the traffic volume reliability by finding the specific events that affect the traffic volume by applying the unstructured data. However, there is a difficult problem replacing with the appropriate data in the present system. To address this problem, in this paper, we show that the accurate traffic amount can be estimated by applying the most similar data through correlation analysis.

Yonghoon Kim, Mokdong Chung

Multi-screen Patterns and Multi-device Experiences in a Multi-screen Ecosystem

With the advent of smart devices, such as smartphone, smart TV, and tablet PC, users own multiple devices and want to use those devices simultaneously or use an appropriate device for their circumstances. In this study, a multi-screen ecosystem is considered, in which users can utilize screens of multiple devices simultaneously to use convergence services and can interact with multiple screens for better user experience. In addition, the service experiences and patterns in a multi-device ecosystem were investigated, in which various smart devices cooperate according to users behavior and needs. Finally, web-based service interaction and migration platform for seamlessly migrating web services in a multi-device ecosystem were discussed.

Geun-Hyung Kim

Software Defined Personal Area Network for Secure and Efficient File Management

As the use of wearable devices grows significantly, it is expected that the Personal Area Network (PAN) will play an important role due to its convenience and efficiency in linking the devices. A file sharing scheme can be employed for secure and efficient management of files stored in the devices. However, wearable devices’ battery consumption may be a problem when using traditional file sharing techniques because the battery capacity is relatively small and the heavy and complex file sharing and retrieval process is demanding. In this paper, we propose a solution to this problem, by applying a Software Defined Network (SDN) to the PAN, where a smart device plays the role of the control layer, and the other wearable devices become members of the infrastructure layer. Using this Software Defined Personal Area Network (SDPAN), the smart device determines how to generate file shares for storage and retrieval for loading rather than requiring each smart device to perform these tasks.

Young-Hoon Park, Kwangman Ko

A Study on Resource Scaling Scheme for Energy Efficiency in Cloud Datacenter

Cloud Data Center (CDC) is growing in popularity as academic and industry hot research spot. With the rapid growth of data centers, thousands of large data centers with lots of computing nodes are established. Accordingly, the energy consumption of the CDC is very high. Also, many of the current research studies have not considered server power state transition and its effect to performance and power consumption. In this paper, we build the resource scaling scheme for energy efficiency in CDCs with considered sleep-mode. And then from evaluation result, we proves that our proposed method is able to efficiently manage the resource and reduce energy consumption.

A-Young Son, Eui-Nam Huh

A Machine Learning Approach to Classification of Case Reports on Adverse Drug Reactions Using Text Mining of Expert Opinions

In this paper, we present a machine-learning approach to classify case reports on adverse drug reactions according to the causal relationship of adverse drug reactions (ADR). For this purpose, the Naïve Bayes classification algorithm is combined with text mining technique, and applied to textual data of expert opinion on ADR case reports in the Korea Adverse Event Reporting System database. The proposed algorithm classifies the case reports into three categories such as possible, probable, and unlikely based on the causal relationship. Our experimental results show that the precision and recall of data belonging to possible is much higher than the other data belonging to probable and unlikely. We believe that our approach can be used not only for signal but also for prediction and prevention of ADRs.

Hyon Hee Kim, Ki Yon Rhew

Multi-scale Surface Curvature Based on Mesh Simplification

The surface curvature of the polygonal model represents a local shape feature of polygonal surface around one surface point. It has been used to analyze the shape of mesh model. In this paper, we present a novel approximation method of multi-scale surface curvature based on mesh simplification, which computes the Gaussian-weighted average of the mean curvature in different sized neighbor regions at a point while changing the resolution of the polygonal mesh. The proposed method was tested on different polygonal models and the experimental results showed that the overall shape features of the polygonal models were represented more clearly by our method.

Jaeyong Lee, Kyong-Ah Kim, Yoo-Joo Choi

The microComponent and Its Extension Patterns for Flexible Reuse of Software Artifacts

Software reuse is a strategic approach for increasing productivity and improving quality of software development. The software reuse can be done with every type of software artifacts like source code, design document and so on. However, software engineers often have difficulty in reusing the artifacts, because finding reusable component is time consuming and hard activity: Software engineers, in many cases, have to retrieve the contents of reusable artifact more than one time, and also find the reusable fragment from whole content. The microComponent is one of solution to support fragment-based reuse of software artifacts. This paper proposes the fundamentals of the microComponent and how to support the reuse of the microComponents. Especially we define extension patterns to enhance the flexibility of the microComponent reuse. Our proposition may give such benefits of reducing the effort and time to find reusable components and increasing the reusability of software artifacts, eventually.

Doohwan Kim, Jang-Eui Hong

Using Code Skeleton Patterns for Open Source Reuse

Reuse has become a very common approach in software development such as open source based development and version upgrades of new product models. However, the difficulty of reusing open source software is that the code is frequently modified; especially modifying its entire control structure makes application development more difficult. This paper proposes a method to improve reusability by reducing the modification of open source software by providing code skeleton for whole structure of application. Our proposing code skeleton approach enables software developers to create the entire structure of developing application, and support method-level reuse of open source software. This approach is capable of developing target application systematically and expeditiously without losing the business logic.

Seungwoo Nam, Doohwan Kim, Jang-Eui Hong

Toward Offline Contents Based Software R&D Support System

This paper proposes key features of contents based offline software R&D support system. The system allows software researchers to describe research contents and define relationships between research contents, also it supports automatic validation of the relationships. In this paper, we discuss the pros and cons of the offline word processors such as MS-Word, LibreOffice Writer and WordPerfect as a base system to build a content based software R&D support system. Then, we propose the key features of the content based software R&D support system based on LibreOffice Writer.

Suntae Kim, Joongi Hong, Seounghan Song, Sangchul Choi, JeongAh Kim, Jae-Young Choi, Young-Hwa Cho

Ubiquitous Authentication and Authorization Mechanism for Enterprise Resources Acquisition

As a result of the massive growth and ubiquity of wireless networks, smart phones have become both a popular and indispensable part of modern life. Many services are offered via smart phone, such as entrance guard systems and mobile wallets. Traditional authentication mechanisms use usernames and passwords to verify user identity, however, in order to ensure sufficiently high security, passwords must be changed regularly. Although some mobile phones have near field communication (NFC) technology, which does away with the need for username and password authentication, NFC only recognizes NFC cards, and does not recognize people. Thus information security is still an issue in such systems. This study combines NFC and biometric identity verification technology to achieve authentication in these situations. In addition, the proposed model combines role-based access control to authorize suitable permissions to users. The proposed model achieves ubiquitous and comprehensive authentication and authorization management for enterprise resources.

Mei-Yu Wu, Chih-Kun Ke, Ming-Ru Lee

Visualization Approach for R&D Monitoring – A Tracking of Research Contents Changes Perspective

Management of Research and Development (R&D) is one of the most important things for efficient R&D. In general, R&D management monitors, analyzes, supports, and controls all the activities that occur during the project. However, most of them are confined to the R&D activities themselves, so it is difficult to grasp the actual research contents during the progress. Therefore, this study proposes a visualization technique that can easily monitor and understand the flow and change of actual contents of R&D, thereby making it possible to utilize it effectively and improve the accurate creation and efficiency of research results.

Jae-Young Choi, Jong-Won Ko, Suntae Kim, Young-Hwa Cho

Hybrid Sensing and Behavior-Aware in Pedestrian Hazard Detection

The advances in multiple types of sensing technology, wireless communication, and context-aware services increase interest in the design and development of pedestrian behavior for hazard detection. This paper focuses on research of the hybrid sensing fusion approach that identifies behavior activities and provides behavior-aware alerts for safety to pedestrians. Hybrid sensing techniques use to integrate data gathered from several sensors and increase the reliability of the algorithm for identifying various activities. The main purpose of this paper is to present the overview of hybrid sensing and behavior-aware to apply for the pedestrian hazard detection.

Svetlana Kim, YongIk Yoon

Internet Articles Classification by Industry Types Based on TF-IDF

In order to understand a specific industry field, people usually look at the financial statements of the companies relevant to the industry field. Financial statements have diverse and numerical information but have past financial states of companies because those are usually quarterly reported. So, needs to timely obtain the current states of an industry field is increasing. Proposed method is focusing on internet articles because they are easy to obtain and updated with new information every day. As a preliminary study of extracting information on industries from internet articles, this paper proposes a method to classify internet articles by industry types. The proposed method in this paper computes importance values of nouns in internet articles based on TF-IDF. Using calculated importance values, proposed method classifies articles by industry types. Through experiments, it is proven that proposed method can achieve high accuracy in industry article classification.

Jonghun Cha, Jee-Hyong Lee

Adaptive Opportunistic Routing over DTMANETS: Proposals and Issues

The convenience of small, cheep, and mobile communication devices such as laptops, cell phones, handheld devices, and mobiles sensor nodes, has popularized mobile ad hoc networks (MANETs). With the convenience, interconnection among these devices introduced new dimensions of challenges for the technology to be used for communication. Such challenges include wireless communication, mobility, and portability. Furthermore, the sparse behavior of nodes in turbulent areas, where connectivity is commonly not possible all the time, resulted in yet another exciting technology known as delay tolerant networks (DTNs). This work is related to the association of opportunistic techniques with different scenarios in which different opportunistic elements of relay nodes, e.g. message storage capacity, territory and velocity are classified according to its usefulness in a given scenario.

Javid Ali, Raja Wasim Ahmad, Tahir Maqsood, Junaid Shuja, Yungwey Chong, Soongohn Kim, KwngMan Ko

Accelerated Purge Processes of Parallel File System on HPC by Using MPI Programming

HPC system usually uses a shared filesystem like Lustre as temporary file storage like scratch directory. It needs automated purge process to remove unused files for maintaining optimal performance of the shared filesystem. However, the purge process in large capacity file system takes much time to search and remove target files. In this paper, accelerated purge processes using MPI are proposed. First, master/slave parallel purge (MSPP) process is the method that master node distributes purge tasks among slave nodes. Second, evenly distributed purge (EDPP) process is the method that all node are involved in purge process that improves load balancing. Experimental results show that the purging time of proposed EDPP method for 1958 GB scratch data has been reduced by 4.04 and 2.47 times, respectively, when it is compared with the results of single node purge (SNP) and MSPP methods.

Min-Woo Kwon, JunWeon Yoon, TaeYoung Hong, ChanYeol Park

Path Privacy Preservation Using Threshold Secret Sharing via Distributed Obfuscators in Directions Search

On directions search, a frequently used mobile service, the path queries of a user may reveal the user private information, e.g., heath status, to service provider. In this paper, we propose a path privacy preservation mechanism utilizing K-threshold secret sharing through distributed obfuscators against directions search provider. A user randomly selects a set of K obfuscators in a pool of obfuscators. The obfuscators have a role in generating obfuscated path queries and querying them to a search server. They sign the query results with their partial secret and return them to the user. The user can verify the signed results and filter the results of the real query. Our proposed mechanism addresses the problem of one point of failure for an obfuscator and collusion attacks. Processing overhead can be controlled by managing the K value of partial secret share or the number of obfuscated queries, or clustering the obfuscated queries.

Mihui Kim

A QoE Based Trustable SDN Framework for IoT Devices in Mobile Edge Computing

The rapid growth of mobile services and Internet of Things (IoT), have caused a severe demand of a management system for Mobile Edge Computing (MEC) where User Equipments (UEs) benefit from high computational power, capacity and communication as well as the offered services by MEC. However, a comprehensive management is required to orchestrate the services and resource in MEC to fairly distribute to UEs with the aim of ensuring the Quality of Experience (QoE). In this paper, we propose a new and trustable framework for MEC management/orchestration system with crucial security and authentication components by which it ensures the delivery of users’ quality of experience.

Hamid Tahaei, Kwangman Ko, Wonjeong Seo, Suchong Joo

Cross-Cultural Touch-Based SNS Interface Design for the Elderly

A study was conducted to identify the relationship between national culture and design preferences of SNS interface features. Value Survey Module (VSM) 2013 was used to measure the cultural dimensions score between countries. Independent t-test and Spearman’s Rho were used to test the hypotheses. This study included a survey with a total of 86 elderly participants, 40 in Korea and 46 in Indonesia. There were significant differences found on the cultural dimensions score. The significant difference was also found on the touch-based SNS interface features rating score and the relationship between the SNS interface features and the underlying cultural dimensions. These results are summarized to provide the ideas for designing touch-based SNS interface for the elderly based on the cultural differences.

Fanny Febriani Susilo, Jung-Ho Lee, Ji-Hyung Park, Jung-Min Park

A Framework for Blockchain Based Secure Smart Green House Farming

The emerging greenhouse technology in agriculture based on Internet of Things (IoT) used for remote monitoring and automation has been rapidly developed. But it still has major concern about security and privacy, due to the large scale of disseminating nature of its network. To overcome these security challenges, we use blockchain which allows the creation of a distributed digital ledger of transactions that is shared among the nodes on IoT network. The main aim of this paper is to provide lightweight blockchain based architecture for smart greenhouse farms to provide security and privacy. Here, IoT devices in greenhouses which act as a blockchain managed centrally to optimize energy consumption have the benefit of private immutable ledgers. In addition, we present a security framework that blends the blockchain technology with IoT devices to provide a secure communication platform in Smart Greenhouse farming.

Akash Suresh Patil, Bayu Adhi Tama, Youngho Park, Kyung-Hyune Rhee

System for the Researcher Map to Promote Convergence Research

As times change, research methods require dramatic changes such as convergence, innovation of research efforts by institutions with similar research goals. Corporations and universities are currently focusing on the development of convergence technologies. This study focused on the role of universities, which can influence the development of such convergence technologies, and developed a researcher map system that provides information to facilitate academic–industrial collaborations for convergence. The system developed in this paper conducted Morpheme analysis, measured keyword similarity, and performed topic modeling based on the titles, keywords, and abstracts of the papers and projects of researchers in different fields to compare semantically similar topics. The results demonstrated higher accuracy than simply calculating the similarity between words.

Sangwon Hwang, Kangwon Seo, Woncheol Ryu, Youngkwang Nam

The Congestion Control Model for Unmanned Aircraft System Traffic Management

With the rapid development of unmanned aerial vehicle (UAV), they have been used in various fields such as logistics systems, agriculture system, and etc. As the consumption of UAV increases, unmanned aircraft traffic management (UTM) system has become necessary. For now, there are few studies on the UTM system for small UAVs. In this paper, we suggest congestion control model in air space by using GPS in real time. We consider the congestion in two ways such as density and traffic jam. When a UAV enters at the range of GCS, GCS figures out whether it is authorized or not. After then, GCS receives the GPS data from each UAVs. With GPS data, GCS calculates density and shows the current situation of density. In order to calculate speed and direction, we use the GPS tracking data. Depending on these data, we can predict the traffic jam. Our proposed method can help to improve navigation system of UAV and to establish UTM system.

Jung-In Choi, Seung-Hyun Seo, Taenam Cho

Towards Recovering Fault Traceability Links by Using Information Retrieval Technique

As the software R&D project progresses, various software artifacts such as software test case, software test descriptions, software source codes, software design descriptions, and software requirements specification will inevitably be produced. When developer performs the system testing, software bugs will be found if there is fault in it. After that, developer will write up an incident report for fault management, and then developer will try to find the artifacts for checking and fixing. Because developer must check the requirement before the fixing and developer has to get the source code including the bug and its test case for fixing and unit testing. Therefore, useful approach should be proposed in order to trace the artifacts from incident report. In this paper, we propose a novel approach for recovering fault traceability links by using IR technique to show that our approach can be a useful solution.

Seungsuk Baek, Jung-Won Lee, Byungjeong Lee

Multi-level Key Establishment with Space-Time Graphs for Delay Tolerant Networks

Protected, low-overhead key establishment is vital to maintain the high level of confidence and security for Delay Tolerant Networks (DTNs). A few works presenting solutions to DTN key establishment have concentrated principally on targeted networking atmospheres. In this work, to deal with the key establishment concern for Bundle Protocol (BP), a time-evolving topology model and two-channel cryptography are formulated to design well-organized and non-interactive multilevel key exchange protocol. A time-evolving model is employed to properly model the periodic and fixed behaviour patterns of space DTNs, and consequently a node can plan when and to whom it should transmit its public key. In the meantime, the application of two-channel cryptography allows DTN nodes to exchange their public keys or revocation position information, with authentication assurance and in a non-interactive approach. This approach facilitates to set up a secure block to maintain BSP, tolerating huge delays, and unanticipated loss of connectivity of space DTNs. The experimental investigation reveals the security enforcement and provides enhancement in performance for DTN network maintenance.

Jinyeong Kang, Inwhee Joe

Defect Management Method for Content-based Document Artifact Test in Software R&D Project

In order to improve the quality of software R&D project, studies on the content-based document artifact test (CbDAT), which considers the heterogeneity of the research phase compared to development phase, have been introduced. The CbDAT manages the research phase by testing the document artifacts which could not be performed by the conventional software test. However, the existing studies on the CbDAT have a limitation in managing because they only consider test planning and test execution. To solve this problem, we analyze the characteristics of the CbDAT and derive six major differences. Based on these, we propose activities that researchers should perform for the defect management. In addition, we present an incident report and defect severity reflecting the characteristics of the CbDAT. The defect management method proposed in this paper enables the researcher to manage defects identified from the CbDAT and to improve the quality of software R&D project.

Dusan Baek, Jong-Hwan Shin, Byungjeong Lee, Jung-Won Lee

An Analysis of Online Learning Tools Based on Participatory Interaction: Focused on an Analysis of the Minerva School Case

The educational environment and related trends have been rapidly changing upon the onset of the 4th industrial revolution. Online learning has been utilized to provide learners with efficient learning opportunities. However, online learning has a problem with the low participation of learners. In order to solve this problem, research on interactive learning tools is being conducted. This study investigates a learning model based on participatory interaction by analyzing the status of advanced cases which utilize interactive learning tools and representative examples. This study provides suggestions for the establishment of a learning model based on participatory interaction. The research results will aid in constructing an online learning environment which could provide highly realistic customized learning, and it has been judged that it will be utilized as basic material in developing LMS for learners’ effective e-learning and in designing a learning model in the future (based on this study).

Dae Hyun Lee, Yen-Woo You, Yong Kim

A New Direction-Based Routing Protocol in WSNs

It is important to extend the network lifetime by using the limited energy of the sensor node efficiently in Wireless Sensor Networks (WSNs). LEACH (Low-Energy Adaptive Clustering Hierarchy) protocol is very well-known as one of most often used cluster based hierarchical routing protocols for WSNs, however, it wastes unnecessary data transmission energy for reverse direction when the sensor node exists between the cluster head and the base station. In this paper, we have proposed D-LEACH (Direction-based LEACH) protocol to reduce such energy wastes by considering the data transmission direction of sensor nodes. Our experiment has shown D-LEACH extends the network life by increasing the efficiency of the data transmission energy usage.

Kyeong Mi Noh, JiSu Park, Jin Gon Shon

Fast Animation Crowds Using GPU Shaders and Motion Capture Data

Animating and rendering of more than 1000 characters in real-time has been a big problem in computer animation research community. In this paper, we introduce the GPU-based technique that are able to animate and render more than thousand of characters at real-time rate where each individual character’s movement is animated using motion capture data. The core part of the proposed technique is to split the whole processes into CPU and GPU bound jobs and separate static data with dynamic data of simulation. All static data are sent to GPU memory once and frequently updated data such as transformation matrices are applied on the GPU using SSBO (Shader Storage Buffer Object). High level path planning, on the other hand, are performed in CPU side so that many complicated algorithms can be easily implemented in the application. Experiments shows that the proposed techniques have real-time performance.

Mankyu Sung

Association-Based Process Integration for Compliance with Core Standards in Development of Medical Software

Several standards for medical software require systematic development to ensure safety and performance. Core standards for medical software include IEC 60601-1, IEC 62034, and ISO 14971. And they present different activities. In this case, standards consist of contents related to development by referring to contents of one another. Therefore, it is difficult for a developer to identify reference-relationships with other standards for complying with one standard. For this, there are rules and studies that assist in the reference-association, but they do not provide the requirements at a view of developer. Therefore, we propose an integrated process to comply with the core standards. The proposed process is defined by analyzing the relationship between development process and so on. Then it includes the requirements and artifacts at each stage of the integrated process. This enables systematic development of medical software by providing the activities and requirements in terms of developer’s view.

DongYeop Kim, Ye-Seul Park, Byungjeong Lee, Jung-Won Lee

Ventricular Arrhythmia Classification Based on High-Order Statistical Features of ECG Signals

One class SVM classification model based on high-order statistical features of ECG signals is proposed. This utilizes distinct features of variance, skewness and kurtosis between normal signals and ventricular arrhythmia ECG signals. The model based on a few simple features motivates immediate treatment for sudden cardiac event and wearable technology in practice. The classification algorithm shows significantly improved performance of 98.9% accuracy in correct classification in the experiment using the MIT-BIH Malignant Ventricular Arrhythmia Database (VFDB). It is expected to be used in real-time electrocardiogram monitoring system in conjunction with ECG measurement part and application part.

Sunghyun Moon, Jungjoon Kim

Compression and Variable-Sized ECC Scheme for the Reliable Flash Memory System

The advances of flash memory process geometry scaledown has lead dramatic capacity increase, however, as a result, endurance is severely degraded resulting in a excessive increase of data errors. To recover the errors occurred during runtime, well known error correction codes (ECC) are integrated into flash memory controller. However, the existing error correction codes have their inherent limits to recover more error data, so more additional schemes should be considered for the higher reliable flash storage systems. In this paper, we propose an architecture for processing combined error correction schemes of compression and variable-sized ECC engine. In the error correction scheme, the original data is compressed with compression module to reduce practical data portion for ECC covered, then, the variable-sized ECC coding is done for the reduced practical data to generate more redundancies. These additional redundancies are stored in the unused area of flash memory page due to compression. The proposed scheme keeps flash memory higher reliable storage systems by dropping down nonlinear increase of bit error rates.

Kijin Kim, Seung-Ho Lim

A Keyword-Based Big Data Analysis for Individualized Health Activity Using Keyword Analysis Technique: A Methodological Approach Using National Health Data

The emergence of Big Data due to spread of the 21st century digital economy can provide a clue to the solution of the problems in our society and economy. Especially, one of the areas where Big Data can be highly useful is the medical and health industry. The development of IT is leading a new era of innovation in the field of medicine as well. Nevertheless, the level of Big Data application in this field is still low as the unstructured data included in Big Data are difficult to search and gather their statistics so that there have been some limits in utilizing it widely. The merit of Big Data is that it can present a variety of meaningful results depending on the data collection and analysis methods. Thus, in this study, a Big Data has been created for text mining by using the crawling technique and analyzed with R Studio, followed by its visualization to devise an individualized health plan with different perspectives.

SangDo Lee, Hoanh-Su Le, Jun-Ho Huh

Reliability and Control Theory: An Integration Approach for Safety Analysis

This paper presents an integrated safety analysis methodology for safety critical systems. In first approach, known as evolutionary safety analysis, we describe system failure models through hierarchical system structure including different safety analysis techniques like Preliminary hazard analysis (PHA), Hazard and operability study (HAZOP), Fault tree analysis (FTA) and Failure mode and effect analysis (FMEA). In second approach, known as revolutionary safety analysis, we combine the results from the first approach for a systematic analysis of scenario based safety control. So far, these two-methodologies seen as two different competing paradigms and have been used separately one for the reliability theory and another for the system and control theory. In this paper, we describe their interrelations and how they can be bridged together for high level of safety. We exemplify our integrated methodology to the development of Green Line Metro System and evaluated the automation via formal verification techniques.

Anit Thapaliya, Gihwon Kwon

An Extended Hierarchical Safety Analysis for Software-Intensive System

Generally, safety analysis is difficult to apply to software that has the characteristic of resulting the wrong system behavior, not as a failure. So many researches continue to relate software safety analysis. This paper presents an extended hierarchical safety analysis method for software-intensive system which combines hierarchical safety analysis and software safety analysis. Failure mode and effect analysis (FMEA), Hazard and operability study (HAZOP) along with Software FMEA (SWFMEA) were applied to perform the safety analysis of model railway system.

Daehui Jeong, Gihwon Kwon

Toward Providing Automatic Program Repair by Utilizing Topic-Based Code Block Similarity

In this paper, we propose the model for automated repair in software fault. Automated patch generation is the most important technique in these days. Genetic Programming (GP) technique is used for automatic program repair, but most of the techniques use just a source code including fault to make initial population. We propose two methods to select similar bug fixing history; using topic modeling and finding similar bugs by using code block similarity.

Youngjun Jeong, Kyeongsic Min, Geunseok Yang, Jung-Won Lee, Byungjeong Lee

Comparing IO Visor and Pcap for Security Inspection of Traced Packets from SmartX Box

With the dawn of distributed cloud computing technology running on hyper-converged box-style hardware, infrastructure operators are facing two challenges of minimizing resource overhead and ensuring infrastructure security. In this paper, we try to compare IO Visor-based and pcap-based packet tracing for security inspection of traced packets from Linux-based hyper-converged SmartX Box. For security inspection, we implement the integration of IO Visor packet tracing with Bro IDS by employing customized scripting and experimentally validate the security inspection performance.

Muhammad Ahmad Rathore, Aris Cahyadi Risdianto, Taekho Nam, JongWon Kim

Building the De-obfuscation Platform Based on LLVM

Following the steady growth of the android mobile market, android application developers apply the obfuscation techniques to hide confidential information to be revealed and to prevent from abnormal approaches. Keeping up this trend, however, the obfuscation technique is also widely used in android malicious codes. Android malicious code developers apply the obfuscation techniques to hide their malignant act and evade anti-virus program, eventually making malicious code reversing engineers spend a lot of time and efforts and adding a huge amount of social cost for the analysis. Due to this reason, the de-obfuscation techniques is getting more and more required to solve this problem. In this paper, we research existing obfuscation and de-obfuscation techniques which currently are applied to the android applications, then suggest the de-obfuscation platform based on LLVM (Low-Level Virtual Machine) to perform de-obfuscation process more efficiently.

JiHun Kim, Kwangman Ko, Jonghee M. Youn

Understanding Automated Continuous Integration for Containerized Smart Energy IoT-Cloud Service

DevOps-based CI/CD (Continuous Integration & Delivery) automation has become important in providing agile and economical services based on micro-services architecture (MSA). In this paper, by taking the example of containerized smart energy IoT-Cloud service, we explain continuous integration with associated testing features for MSA-based service realization.

Chorwon Kim, Seungryong Kim, JongWon Kim

Comparing the Effectiveness of SFMEA and STPA in Software-Intensive Railway Level Crossing System

The complexity of software-intensive systems is a challenge for software developers in choosing the optimal method from hundreds safety analysis methods. This paper proposed a comparison between two common safety analysis techniques: Software Failure Mode and Effect Analysis (SFMEA) and System Theoretic Process Analysis (STPA). The comparison is based on the results of both methods applied in one case study: Level Crossing system. The comparison results are useful for safety analysts in choosing appropriate techniques.

Tung La-Ngoc, Gihwon Kwon

Vision-Based Humanoid Robot Control Using FIR Filter

In this paper, we propose a novel vision-based humanoid control method and visual tracking based on constant velocity (CV) model using the finite impulse response (FIR) filter. The proposed method has robust performance even if a sampling time or noise information is inaccurate. Furthermore, even when the movement of the detected ball or the ambient illuminance changes suddenly, the proposed method shows robust performance. The robust performance of the proposed method is verified through experimental results.

Kwan Soo Kim, Hyun Ho Kang, Sung Hyun You, Choon Ki Ahn

Prototype Implementation of Site Visibility Framework Employing IO Visor-Based Packet Tracing

With the growing popularity of cloud-leveraged infrastructures and services, the clustered operation of multiple physical and virtualized boxes is rapidly increasing in a single site. It is however very hard to monitor and control the targeted site for security management when there exist a number of inter-connected boxes. To address this issue, in this paper, we utilize IO Visor-based packet tracing to inspect and collect the packets from multiple boxes in the site. Also, we introduce the concept of Site Visibility Framework by leveraging IO Visor-based packet tracing, which supports packet-level monitoring for security management of inter-connected boxes in the site.

Taekho Nam, JongWon Kim

ICN Based Disaster Area Network Platform

Information centric networking (ICN) is a next generation network technology that is expected to replace the current Internet. The fundamental concept in ICN is the access of named data as a principal network service. In this paper, we propose the idea of an ICN based disaster area network platform that provides the functionality of application services, including management and workflow, for disaster information. The platform bridges between disaster-affected and non-affected areas using unmanned aerial vehicle equipped with ICN router, and also supports cooperation between the current Internet and ICN.

Masashi Katsumata

Single-Camera Vision-Based Vein Biometric Authentication and Heart Rate Monitoring via Infrared Imaging Analysis

In this study, a feasibility test is performed to evaluate simultaneous heart rate measurement and individual identification via a single device. Thus, we have designed a novel device, comprising a modified webcam and 660 and 940 nm LEDs, as based on the principles of a conventional blood flow measurement sensor. To confirm the feasibility, we captured three types of images via respective employment of the following: 660 nm LEDs only, 940 nm LEDs only, and both 660 and 940 nm LEDs. A PPG (photoplethysmography) signal is acquired as the images are captured; the output is implemented as ground-truth data. Experimental results showed that the image analysis-based heart rate signal yields a pattern identical to that of a PPG signal. Additionally, acquired finger-vein image visibility and resolution is sufficient to perform finger-vein recognition. Testing the system using more than 100 subjects with variable health statuses confirmed that our proposed concept can be implemented as an effective heart rate monitoring system. The proposed method has the potential to significantly increase the efficiency of individual health information management.

Jae Hyun Han, Jinman Kim, Eui Chul Lee

Analysis of Agenda Prediction According to Big Data Based Creative Education Performance Factors

In the present study, to derive future mid/long-term development directions and agendas according to the outcomes of South Korean creativity education policies that have been steadily implemented thus far, opinion mining analyses were conducted utilizing educational data. With regard to analysis methods, creativity education related unstructured data were collected, linkage analysis based higher education policy keywords were extracted, and opinion mining analyses were conducted through the extracted keywords. From the analyzed results, we derived educational systems that can be very important for future development of creativity education and performance factors through the positive and negative data on the educational policies that are currently being implemented. The outcomes of the present study will be a solution that can be utilized hereafter in preparing direction points according to domestic and foreign educational policies.

Ji-Hoon Seo, EunMi Cho, Kil-Hong Joo

Multimedia Design Approaches by Just Noticeable Difference (JND) of Audiovisual Modalities

Multimodalities have attracted much interest in designing effective multimedia contents and services by applying sensory information. However, sensory experience from different modalities have been less understood in the multimedia contents design. Therefore, the optimal sensory experience is required to be considered to give impact on multimedia contents. This study is to determine the normalized perception for equalizing sensory difference between audio and visual modality. They were estimated by Just noticeable differences (JND) of audio and visual modalities and combination of both JNDs. JND for sound pressure and brightness were measured or estimated by curve fitting to calculate normalized just noticeable difference (NJND). NJND showed the intensity of sound pressure and brightness for the sensory balance and the sensory dominance.

Suhhee Yoo, Mincheol Whang

Usability Improvement of Life-Logging Contents Based on Gamification Factors

Recently, the application of a gamification factor to application contents has been initiated, but it is not actively applied in life-logging applications because it focuses on information recording and data visualization. The purpose of this paper is to verify that the usage rate is improved when the gamification element is applied to a life-logging application. We individually classified them into groups, applying gamification factors of ‘Acting’, ‘Achievement’ and ‘Social’ for the life-logging application, along with a ‘3 factors’ group applying all three elements to carry out the experiment. As to the results of the experiment, we confirmed significance in that the usage rate of the group applying gamification factors was 6.4 times or more than that of the control group. Based on the experiment results, it is expected that it will serve as a basis for providing enjoyable behavior and application usage rates beyond the simple function of user records by applying gamification factors to life-logging applications effectively.

Sojung Kwak, Jieun Kwon

The Relationships Between Behavioral Patterns and Emotions in Daily Life

Emotions have been recognized from physiological and behavioral responses, however, in daily life these methods are less practical due to the measurement burden. This study was to minimize the measurement burden by using smartphones and to determine the behavioral patterns relevant to daily emotions through the global positioning system (GPS) locations. Seven participants (5 males) were asked to carry their smartphones and evaluate subjective emotions for six weeks. The participants’ GPS locations were measured with their smartphones and then analyzed to determine their behavioral patterns. The emotions were categorized into valence and arousal dimensions, and the behavioral patterns were tested by the Kruskal-Wallis method. As a result, the valence dimension implied significant behavioral patterns such as location variance (p = .006), number of cluster (p = .015), and entropy (p = .044). The arousal dimension implied significant behavioral patterns such as location variance (p = .003), circadian movement (p = .008), and transition time (p = .016). These behavioral patterns are expected to be useful in recognizing emotions in daily life.

Hyunwoo Lee, Ayoung Cho, Youseop Jo, Mincheol Whang

Automated Verification Method of Korean Word Handwriting Using Geometric Feature

Handwriting is a behavioral characteristic of an individual, and is used as a biometric method for identity verification. The handwriting comparison can not be trusted because the handwriting similarity is judged by the subjective criteria of the expert. In this paper, to compare two images written with the same Korean word, some digital image processing techniques are used to detect geometric feature points such as corner, ending, and bifurcation points of a stroke. Then, adjacent redundant feature points is removed. Word images of different sizes are normalized by multiplying the coordinate values ​​of the feature points by the magnification of the image. The similarity between two sets of feature points is measured through modified Hausdorff distance. In the case of genuine matching, it is confirmed that there is a correlation between the number of feature points and distnace. Therefore, the calculated distance is normalized by dividing the number of feature points by the number of feature points. Experimental results show the possibility of automated handwriting verification by showing 37% EER.

Woohyuk Jang, Sehee Kim, Yoonkyoung Kim, Eui Chul Lee

Correlation Analysis Between Environmental Sound and Human Emotion

In daily life, as people are most exposed to the surrounding environment, accordingly humans are greatly affected as their emotion depending on the surrounding visual and spatial information. In this paper, we propose an analysis method of how the sound information such as amplitude and frequency in the surrounding environment can affect to human emotions by adopting sound features among the visual and spatial information. For the experiments, a total of 1,500 video clips of surrounding environment are acquired by 15 subjects using the camera built in smartphone. Also, the subjective evaluation were performed after taking the video. Two features such as amplitude and frequency of sound data were extracted. Then, we designed a fully connected SVR inference networks in which the data were divided into two sets such as training and test. The extracted two dimensional features were SVR trained by corresponding with the subjective evaluation scores for pleasant and arousal levels. As a result, we confirmed that the estimated two-dimensional emotions were similar with the subjective evaluated ones in which the errors were about the pleasant level (−1 ~ +1) of 0.27 and the arousal level (−1 ~ +1) of 0.32, respectively.

Min Woo Park, Hyeonsang Hwang, Eui Chul Lee

Image-Based Malware Classification Using Convolutional Neural Network

In this paper, a malware analysis method that analyzes images learned by artificial intelligence deep learning to enable protection of big data by quickly detecting malware, including ransomware, is proposed. First, more than 2,400 datasets frequently used by malware are analyzed to learn and image data with a convolutional neural network. Data are then converted into an abstract image graph and parts of the graph extracted to find the group where malware exist. Through comparative analysis between the extracted subsets, the degree of similarity between these malware is analyzed experimentally. Fast extraction is achieved by using deep learning. Experimental results obtained indicate that use of artificial intelligence deep learning can enable fast and accurate malware detection by classifying malware through imaging.

Hae-Jung Kim

Classification of Web Content by Category Generation in Social Life Logging

Web content is consumed at anytime and anywhere through mobile devices. Consumption behavior has been affected by its own emotional content. Web content has been categorized by article’s topic and its emotion has been determined by article’s nuance. This study is to determine category and emotion of web content. The Automatic Content Categorization System (ACCS) has been developed to crawl the texts from web page and to separate texts into morpheme using natural language processing (NLP). Finally, web content was classified into category and emotion by document similarity. The main contribution of this study is to provide fixed categories and 28 emotions to classify web content for analyzing consumption behavior of web content.

Youngho Jo, Heajin Kim, Hana Lee, Mincheol Whang

Intrusion Detection in High-Speed Big Data Networks: A Comprehensive Approach

In network intrusion detection research, two characteristics are generally considered vital to build efficient intrusion detection systems (IDSs) namely, optimal feature selection technique and robust classification schemes. However, an emergence of sophisticated network attacks and the advent of big data concepts in anomaly detection domain require the need to address two more significant aspects. They are concerned with employing appropriate big data computing framework and utilizing contemporary dataset to deal with ongoing advancements. Based on this need, we present a comprehensive approach to build an efficient IDS with the aim to strengthen academic anomaly detection research in real-world operational environments. The proposed system is a representative of the following four characteristics: It (i) performs optimal feature selection using branch-and-bound algorithm; (ii) employs logistic regression for classification; (iii) introduces bulk synchronous parallel processing to handle computational requirements of large-scale networks; and (iv) utilizes real-time contemporary dataset named ISCX-UNB to validate its efficacy.

Kamran Siddique, Zahid Akhtar, Yangwoo Kim

Embodied Emotion Recognition System

The relationships among body, brain, and environment have been important to recognize emotions according to the embodied emotion theory. Its relationships should be analyzed in real time because the relationships have been changed every moment. Therefore, this study has developed a system that automatically analyzes the complicated interactions based on personal data by path analysis in real time. The system has three phases: First, data have been collected with a wearable and smartphone device to measure photoplethysmogram (PPG), global positioning system (GPS) locations, and ambient noise. Second, features have been extracted by preprocessing. Finally, interactions have been determined by calculating the directed dependencies with path analysis. As a result, the relationships are presented in directed graphs form. This system is expected to be a useful platform to recognize and to predict human behavior based on the interactions among body response, brain response, and environment.

Ayoung Cho, Hyunwoo Lee, Hyeonsang Hwang, Youseop Jo, Mincheol Whang

Patterns of Cardiovascular and Behavioral Movements in Life-Logging According to Social Emotions

The purpose of this study was to determine the cardiovascular and behavioral patterns to develop a new algorithm of emotion recognition system through only behavioral patterns. However, to achieve this we must compare the features with both cardiovascular responses and subjective evaluations. Seven students were asked to wear PPG sensors and carry their smartphones to track locations and periodically evaluate subjective emotions. The social emotions were categorized into mutuality and sociality dimensions. As a result, in sociality, cardiovascular features implied significant patterns in 8 cardiovascular features (p < 0.01). In mutuality, significant patterns were implied only in total power (p < 0.01). Additionally, results for sociality in behavioral features implied significant patterns in transition time and total distance (p < 0.01). Cardiovascular and behavioral patterns are two factors that can determine the physiological effects of individuals according to emotions.

Hana Lee, Youngho Jo, Heajin Kim, Mincheol Whang

Deep Representation of Raw Traffic Data: An Embed-and-Aggregate Framework for High-Level Traffic Analysis

In Intelligent Transportation Systems (ITS), it is widely used to extract a fixed-size feature vector from raw traffic data for high-level traffic analysis. In several existing works, the statistical approach has been used for extracting feature vectors, which directly extracts features by averaging speed or travel time of each vehicle. However, we can achieve a better representation by taking advantage of state-of-the-art machine learning algorithms instead of the statistical approach. In this paper, we propose a two-phase framework named embed-and-aggregate framework for extracting features from raw traffic data, and a feature extraction algorithm (Traffic2Vec) based on our framework exploiting state-of-the-art machine learning algorithms such as deep learning. We also implement a traffic flow prediction system based on Traffic2Vec as a proof-of-concept. We conducted experiments to evaluate the applicability of the proposed algorithm, and show its superior performance in comparison with the prediction system based on the statistical feature extraction method.

Woosung Choi, Jonghyeon Min, Taemin Lee, Kyeongseok Hyun, Taehyung Lim, Soonyoung Jung

A Study on Traffic Signal Waiting Model Using Queuing Theory

This study analyzed the traffic situation by focusing on the characteristics of the customer after gaining idea from the ghost congestion phenomenon which is still congested even if the cause disappears once the congestion starts. We have developed a model that can quantitatively analyze queuing theory to apply to traffic situation. It is difficult to apply the queuing theory because the number of cars leaving the unit per hour varies greatly depending on the situation. Therefore, this study designed the model reflecting the variables that can occur in the actual traffic situation and obtained the throughput according to the situation. Simulation was carried out to verify the validity of the model. The concept of individual throughput for each customer was introduced, and the relation with the throughput was expressed through formulas. This research is useful in studying systems that have to reflect the characteristics of individual customers, or where interactions among customers play a large part in queuing.

JoongHoon Lee, HyuckJoong Yoon, Tae-Sun Chung

Service Aware Orchestration for Dynamic Network Slicing in 5G Networks

In 5G networks, Network slicing enables the operator to provide customized networks by slicing one physical network into multiple, virtual, end-to-end networks, referred to network slices. Each network slice can be defined according to different requirements on functionality, performance and specific users. Based on the concept of network slicing, a single device connects to multiple network slices simultaneously, are described in order to indicate the necessity of multiple network slicing. It is also specified necessary functions in order to support how to select a proper network slice, how to manage the mobility of different session in a same device. Finally the service aware orchestration is proposed in terms of network slice creation, management and deletion.

Jeongyun Kim

Improved Schedulability Analysis of Fixed-Priority for Mixed-Criticality Real-Time Multiprocessor Systems

In real-time systems, timing guarantees on real-time tasks are the most important requirement. The requirement, however, causes inefficient resource usage due to over-estimated (but safe) task execution time, which brings the advent of the concept of Mixed-Criticality (MC). In this paper, we improve an existing schedulability analysis of Fixed-Priority (FP) scheduling on MC systems. Our evaluation results demonstrate the effectiveness of our analysis in terms of schedulability.

Namyong Jung, Jinkyu Lee

Analysis of the Elements of Future Development of Korean Style Software Education Through the Opinion Mining Technique

The present study analyzed the elements of future-oriented development of Korean style-based software education for more effective operation centering on the software education that will be implemented as a regular curriculum subject from 2018 in South Korea. As for the analysis methods, unstructured data collected from the Web were managed in a big data store and trend analyses were conducted through the frequency counts of words utilizing the pretreatment technique. Based on the results of the present study, the proportion of positive elements has been higher thus far. However, the proportion of negative elements has been increasing over time. Therefore, it could be seen that for long-term development of software education, rather than coding centered education, the proportion of the learning of the ability to think computing would act as an important variable for development factors.

Ji-Hoon Seo, Kil-Hong Joo

Comparison of 2D&3D Performances of Facial Feature Analysis Using RGB-D Vision Sensor

This study was conducted to experimentally identify a better method of facial expression recognition using the AAM (Active Appearance Model), a two-dimensional image-based method and three-dimensional depth information method. Experiments were performed by com-paring facial feature points for happy facial expressions and neutral facial expressions, and analyzed the 5000 sets of 2D and 3D facial feature point data of university student subjects. As a result of the analysis, it is confirmed that the same facial feature vector change is more clearly distinguished when using the 2D image based method than the 3D depth information based method. It is confirmed that this phenomenon is caused by the fundamental problem of structured light type RGB-D camera which causes error up to 15 mm at 1 m depth. Consequently, the 3D method can be advantageous when facial expression recognition through AAM is frequent in the depth direction or facial pose variation is large, but 2D method has excellent performance for accurate facial recognition in a static situation respectively.

Kunyoung Lee, Eui Chul Lee

The Two Dimensional Model of Social Emotion Based on Social Life Logging

Social emotion is interactive emotion in human relation. SNS (Social Network Service) has been active domain in expressing emotion developing social emotional connection. However, social emotion has less been determined to be related to emotion expression. Therefore, this study is to determine the dimensional representatives of social emotion by analyzing cementics of emotional words in SNS domain. 24 SNS has been selected for analysis in this study. 1,153 emotional adjectives was extracted and their frequencies of usages were evaluated. They were reduced to 85. 56 emotional adjectives were finally determined by test of goodness-of-fit. Representative words were defined by cicular ordering task and category sort task and their coordinates were determined in 2 dimensional domain of emotional words. 56 emotional adjectives were mapping into the 2 dimensional coordinates from location of representative words. The orthogonal dimensions, as the results, was work centric and leisure centric. The others were relationship centric and isolation centric. The social emotion in SNS has been characterized from activities of both work and relationship.

Heajin Kim, Youngho Jo, Hana Lee, Mincheol Whang

Design of Zigbee-BLE Gateway Direct Communication System for Smart Home Environment

With the recent introduction of network technology to home appliances, there is a growing interest in smart homes that provide convenience in life using home gateways. However, as smart devices are not equipped with Zigbee modules, a server is required to access home appliances through their Zigbee modules. Therefore, in this study, a Zigbee-Bluetooth low energy (BLE) gateway direct communication system was designed, which is capable of direct access between smart devices and Zigbee-based home appliances through communication between Zigbee and BLE without a server.

Jae-Sung Shim, Hyung-Joon Kim, Nam-Uk Lee, Seok-Cheon Park

Design of Automatic Source Code Generation Based on User Pattern Definition

Software engineering has been established with various development methodologies and software development process models, which have enabled increases in software production efficiency and improvements on product quality. However, the time taken to develop software and input human resources are the main causes of cost increases. Thus, the project will be delayed as the time taken to complete iterative tasks increases. In this regard, this paper designs an automatic source code generation system using meta data based on user patter definition in order to resolve the problem.

Seung-Su Yang, Hyung-Joon Kim, Nam-Uk Lee, Seok-Cheon Park

Design of TDD-Based Automation System for Android Application Test Automation

The Android platform does not support interoperability among different platforms, and this causes difficulty for developers because they must test applications on different platforms and devices. To solve this problem, this study proposes a test automation system applying test driven development (TDD) that allows automatic performance of repetitive tests. To design the proposed system, test case generation was automated using Annotation.

Min-Hyung Park, Hyung-Joon Kim, Young-Hwan Jang, Seok-Cheon Park

Convolutional Neural Network Based Serial Number Recognition Method for Indian Rupee Banknotes

The recognition of the banknote serial number, which constitutes important data used for various purposes is one of the important functions of banknote counters. However, traditional character recognition methods are limited in terms of speed and performance of serial number recognition. Therefore, in this paper, we propose a character extraction method based on the aspect ratio of banknotes and a character recognition method based on a convolutional neural network (CNN). For character extraction, de-skewing was performed first. Then, the serial number was estimated on the basis of the aspect ratio of the banknote. Further, we designed four types of CNN-based neural networks for character recognition and adopted the most appropriate neural network. Subsequently, we confirmed that the average recognition performance per character for each neural network was 99.85%.

Unsoo Jang, Eui Chul Lee

Emotion Recognition Through Cardiovascular Response in Daily Life Using KNN Classifier

Emotion in daily life is difficult to recognize due to disadvantageous of continuous measurement. This study was to develop the method for recognizing daily emotion from a measurement of daily cardiovascular response by using the developed wireless sensor. Seven subjects assessed subjective emotions based on Russell’s emotional circumplex model every 3 h wearing a photo-plethysmography (PPG) sensor. The heart rate variability (HRV) according to two emotional dimensions were tested by the Kruskal-Wallis test. Significant parameters of them were determined to be distinguished among emotions and were applied to recognize emotions using the K-Nearest Neighbor (KNN) algorithm. The arousal and valence were recognized with respective 88.2% and 56.2% accuracy. The methods in this study is extended to monitor and recognized in industrial domain and health care domain requiring recognition of long-term emotion.

Youseop Jo, Hyunwoo Lee, Ayoung Cho, Mincheol Whang

Location Privacy for HIP Based Internet of Things

The current IP address scheme, which uses both concept of identifier and locator, has problems such as scalability problems, service interruption, multi-homing support, and ease of exposure of privacy information. In order to solve these problems, the host identity protocol (HIP) was developed to achieve an identifier-locator split by introducing an additional namespace between network and transport layer that provides stable cryptographic host identifiers using public keys. However, the privacy problem due to the exposure of the location information and the tracking and linking of both information are also problems to be solved. We propose a location privacy mechanism for HIP-based network architecture with Rendezvous server. It provides location privacy to the other nodes using pseudonymous IP. As a result, it satisfies location privacy and unlinkability.

Kyung Choi, Mihui Kim

Ubiquitous Learning and Digital Badges in the Age of Hyper-connectivity

Ubiquitous learning (U-learning) in the age of hyper-connectivity offers a new learning space that is shifting the dynamics of the formal to informal learning continuum, and therefore encouraging industries to explore alternative preferences for learning and credentialing in the 21st century. Lately, the uses of digital badges to credential learning have been amplified in schools and businesses through many innovative means. Yet, badge application is an open-textured concept where no single static application exists. This paper discusses salient characteristics of badges from a pedagogical standpoint and discusses exemplary case studies of badge use in the education and business sectors. This paper contributes to the discourse on the badge use as a viable mechanism for learning credentials in a U-learning society. The paper also identifies the current limitations of badges where the authors recommend that badges be regarded as an additional instrument in the scaffolding of learner understanding of crucial material, rather than it being intended to replace traditional formats of learning validation.

Yoonil Auh, Heejung Raina Sim

Private Data Protection of Android Application

Many Android applications store sensitive information such as account information, debit/credit card details. The exposed information can be accessed by attackers. They can pretend to be the data owners in cyber space with a duplicate of the information. In this study, we propose a new framework to protect Android applications against the data duplication attack. The proposed framework is practical because it can be applied to all of Android applications and has an inconsiderable overhead contributing to security of Android.

Jinseong Kim, Im Y. Jung

Erratum to: An Efficient Clustering Technique for Unstructured Data Utilizing Latent Semantic Analysis

Yonghoon Kim, Mokdong Chung

Backmatter

Weitere Informationen