Advanced Multimedia and Ubiquitous Engineering

Internet users usually subscribe to a few online services. Remembering a different password for each service becomes a burden and a challenge for some. As a result, many Internet users frequently use the same password for multiple accounts. This kind of practice is risky since each service has a different security level. For example, an online community site has a weaker security measure than an online bank site. If an attacker has compromised a lower security service and obtained the user’s password, the attacker may be able to identify other accounts and use the stolen password. Therefore, reusing passwords becomes a security risk, and is not generally recommended. This paper tries to mitigate the risk of reusing an identical password for multiple accounts by implementing a single password authentication protocol. The proposed protocol does not expose the user’s password in the event of the server or the communication line has been breached.

This paper describes scheduling and processing of congested packets in massive Machine Type Communications (MTC). When there are too many uplink packets, it may cause congestions in the Radio Access Network (RAN) and the Evolved Packet Core (EPC). To solve this problem, we propose a Critical Random Early Detection (CRED) method to compute the current average queue length for each process to determine whether a packet needs to be discarded or not to prevent from congestions. The proposed method is evaluated through NS-2 simulations.

One of the key issues concerning IT systems is Information Security Management. Among the security objectives in the ISO/IEC 27002:2013 standard refers to information security awareness, education and training. In this area there are many important aspects but in this paper authors focus on people, their knowledge and their security awareness. Authors introduce a model that could illustrate organization members, their relations and knowledge about security. Results of simulations can be used to create plans of training to increase their security awareness. Finally authors present few cases where different strategies of teaching people are tested and the analysis is presented. If knowledge does not change under the influence of co-workers, it is better to train those with smallest knowledge.

The expansion of cloud service has brought about the change of communication method between computer and user. Therefore, the significance of cloud data management system has increased. However, the current cloud data exchange method lacks in consideration regarding the utilization of big data. This thesis proposes a framework for cloud data management system considering the big data processing. For this, the thesis used CDMI standard structure to conduct separation and compression of cloud data to propose an efficient data management system.

It could be said that AI (Artificial Intelligence) is the most prominent field within the short history of computers and it is recognized as an important element deciding upon future human life. The study aims to explore the exact concept of AI and environments used in e-learning. Through this, the study aims to review learning combined with Artificial Intelligence (AI), thus, Internet self-directed learning which is recently receiving much interest and explore actual fields where AI is currently applied. As a result, the study aims to suggest the base data for providing students a more efficient and abundant education environment using artificial intelligence.

Contemporary society has been seeing rapid progression IT since the information revolution that was launched with the invention of the computer in the 20th century. Everyone living in contemporary society is affected directly or indirectly by IT in a wide variety of areas including medical services, transportation, fashion, business management, administrative management and art. With the rapid penetration of the internet, information in the cyber world is also on a dramatic increase [1]. Since the economic value and importance of such information are also on the increase, there have also been much increase in the risk factors that target highly valued information. Even though the risk factors are on the rise in tandem with the progress made in IT that make our life more convenient, there is a lack of awareness of information security among the public. Therefore, this study seeks to explain the concept of computer viruses and how they developed, the risk posed by ransomware that have newly emerged as a threat and citizens’ ransomware that is in contrast to that. It also seeks to analyze the results of a survey on the perception and understanding of the importance of cyber information security, thereby promoting the risks of ransomware and raising awareness. In doing so, the study seeks to present a basic set of data that can serve as reference for future measures to be taken.

In this paper, a method of quantitatively extracting human emotions by analyzing the surrounding environment images obtained through smartphone cameras in real time is proposed. In the area psychology, it has been known that visual elements such as colors and complexity affect human emotions. Based on the foregoing, we developed an application for the extraction of emotions in real time using the colors and spatial complexity of images obtained through android smart phones. Among the color components of images, hue components that indicate colors were extracted as color elements and the spatial complexity was extracted through the quantities of the high frequency and low frequency components out of the frequency components of images, respectively.

Fingerprint recognition is a biometric method. Recently, attempts for spoofing of fingerprint recognition systems through fake fingerprints have been frequently reported. Most existing fake fingerprint detection methods require either additional sensors or complicated calculations. In the present study, a new fake fingerprint detection method implemented through the combinations of six simple statistical moment features is proposed. The six statistical moments mean the deviation, variance, skewness, kurtosis, hyperskewness, and hyperflatness. Average brightness, standard deviation, and differential feature are additionally used. The multi-dimensional features were combined through the Support Vector Machine. According to the results of experiments, the proposed method showed classification accuracy of about 98%.

In recent years, the development of unmanned aerial vehicles (UAVs) has increased significantly and they are currently used in various fields and applications. In some applications, multiple UAVs need to be cooperated to accomplish tasks, because a single UAV is not sufficient. However, even when multiple UAVs are used, their autonomous control systems are not perfect, which leads to collisions between the UAVs. In this paper, we propose a path planning method for collision avoidance of UAVs, when multiple UAVs are controlled using a ground control system. Furthermore, using this method, the UAVs have less likelihood to be in a close encounter with obstacles, and collisions are avoided.

Unmanned Aerial Vehicles (UAVs) have been utilized in various applications in many fields in recent years. The paths the pilots flew can be measured and collected to be utilized to create routes for autonomous flight. However, there is a problem in that GPS errors result in the path being irregularly represented. The measured path can be optimized by using the Douglas-Peucker algorithm. Our research led to the proposal of a method to optimize this path by applying the Douglas-Peucker algorithm, which has been shown to be suitable for a two-dimensional path, in three-dimensional space. Optimization of the 3D path by the proposed method was possible by deleting unnecessary points from the three-dimensional space. Thus, the flight paths that were measured and collected can be utilized to define the autonomous flight path.

As the amount of encrypted network traffic on enterprise networks increases steadily, the problem of malicious acts encrypted to bypass security devices has emerged. Previous studies analyzed the encrypted network traffic by changing the network traffic or communication flow between the encrypted communications to analyze such encrypted malicious behavior. However, there are limitations to the existing methods because they require additional prior-data or additional network configurations in order to analyze the encrypted network traffic. In this paper, we propose a system to decrypt secure socket layer network traffic to analyze the encrypted network traffic in the enterprise network environment. The proposed system can be used to analyze encrypted network traffic in order to detect malicious activity and corporate information leaks.

A considerable amount of spam that occur each year can cause the financial damage as well as mental harm to the recipient. This is a serious problem in society. In this paper, we analyze properties of SMS spam in mobile phones to establish a method for effectively blocking SMS spam. As a result, SMS spam can be seen that the surge in the amount shipped during a specific time period. Also, we could find the frequently included word on spam and we could identify spammer that sent smishing messages frequently by comparing several spammers.

As the number of patches in a patch management system increases due to software updates and security issues arise in the existing patch management system, a more efficient patch management system with reinforced security is required. Additionally, existing patch management systems must be improved, as they perform patch collection inefficiently and their patch integrity verification schemes are simple. In this paper, we propose an automatic patch management system with improved security, enhanced patch collection efficiency, and reinforced verification of patch integrity that automatically collects patches through patch sites.

Smartphones, which debuted in the form of personal digital assistants (PDAs) in the late 1990s, have evolved continuously. However, hardware features and the external environment restrict their use, making it difficult to ensure high interactivity. The structure and usage of mobile applications are also becoming increasingly complex and it is often found to be difficult to understand the user interface (UI). These user environments and conditions inhibit the smooth interaction of the user with the application. This is expected to negatively affect the user’s intention to use the applications eventually. However, past studies on information systems have not shown much interest in the impact of smartphone UI designs on the attitudes and behaviors of users. Thus, this study attempted to empirically explore the impact that UI application designs have on the behavior intentions of the users utilizing the application. This study specifically looked at the following aspects of UI design: simplicity and consistency. The data was collected through a survey and structural equation modeling (SEM) was employed for the analysis. The results showed that these attributes have a significant effect on the interaction as well as a positive impact on the intention to use the application.

Recently, biometrics technology has been receiving attention as means of personal authentication in smartphone environment. Fingerprint recognition is generally contained in newest smartphones and other biometric methods such as iris recognition are receiving attention. However, these methods have a problem of being not applicable to existing smartphones because additional devices such as infrared cameras or sensors should be included. To solve this problem, in the present paper, a new biometric method using features on the rear of the thumb is proposed. The similarity between enrolled thumb images and input thumb images is measured through the SIFT (Scale Invariant Feature Transform) method. Through feasibility tests, it could be identified that the proposed method could recognize the thumb with an accuracy level of approximately 99.94%.

Image classification is helpful for searching and image retrieval in terms of corresponding to the preference of users. However previous works did not consider human emotion but perform the retrieval by using keywords or objects in image. In the field of color psychology, the color has been proven that an impact on the human emotion. Also, visual complexity such as spatial frequency affects to human emotion. In this paper, a new image classification method is proposed for analyzing the relationship between image components such as color and spatial frequency and human emotion. We collected totally 391 images which contained the three different kinds of scene categories such as natural scene, campus scene, and human made scene images from the public image database. Consequently, we confirmed that image can be reasonably classified by using the color and spatial frequency in terms of human emotion.

Recently, human robot interaction technique is limelight in which controlling robot by using natural user interface such as using body gesture, eye tracking or human voice recognition. In this paper, hand gesture based robot controlling method is proposed. In the present study, a new natural user interface model is hierachically designed. The hand gesture is recognized based on Kinect V2. The recognized gesture information is send to the robot controlling robot through Bluetooth 4.0 interface. The developed robot controlling interface can be adopted into the field of disaster area in order to restore or surveillance the area of human inavailable. Experimental result showed that the proposed method can be effectively used for controlling robot.

Applications using parts of a user’s body as an interface are increasing in number. In an indoor environment the location of a user can be measured using a beacon, and the direction of movement can be estimated from their location as measured by the beacon; however, a limitation exists when estimating the direction in which the user is facing. This paper proposes a method for estimating the direction of the user’s body using two beacons. It verifies experimentally that the proposed two-beacon method can correctly estimate the direction of a user’s body, with the error in the user direction reduced by 74.13% when compared to an existing method.

In this paper, we propose a user selection based backpropagation method for siamese networks which we will use as visual filters in mobile contents. The loss function used to train the network is affected by the user’s interaction which also affects the update of the weights in the network such that the visual filter takes subjective similarities into account. As a specific application of the proposed algorithm, we expect that the visual filter can be applied for mobile services which provides the user with 3D visual products appearing above the phone. The products appearing in sequence to the user are those which have similar appearances to that selected by the user where the subjective similarity is also taken into account.

This paper stresses the importance of monitoring indoor environment quality (IEQ) in a bedroom environment for the pre-elderly, by comparing data on sleep behavior with the corresponding IEQ variables. Although concerns regarding the influence of IEQ on residents’ health are increasing, solutions for optimizing bedroom environments have not yet been determined. In particular, for the elderly and pre-elderly, optimization of the indoor environment may relieve the symptoms of chronic disease and degradation of their physical abilities. This study presents a method for obtaining basic data to create an optimized bedroom environment for the aging society.

Differential swelling of spatially patterned gel sheets offers an indirect avenue to rational understanding how things in nature grows as time elapses. In the present study, we demonstrate swelling-driven deformation of spatially designed gel sheets sensitively relying on film thickness and pattern dimension. Through chemical copolymerization of poly (N-isopropyl acrylamide) (PNIPAm) with pendent benzophenone UV-crosslinkers, photo-crosslinkable hydrogels were prepared. Various kinds of spatially designed features with different equilibrium degrees of swelling are created by grayscale and halftone gel lithography techniques, wherein time-sequential UV exposures with a number of photo-masks, thus selectively embedding densely-crosslinked features into a lightly-crosslinked area. Deformation of the photo-patterned gel sheets by energy competition between the regions strongly depends on a characteristic dimension and sheet thickness, providing fruitful information on the contrast in modulus between the regions.

Recently usage of multiple devices including mobile devices, TVs, OTTs, game consoles, have been extremely increased, allowing multi tasks to users. In addition, the rapid increase of connectivity between the devices has occurred. However, the issues of limited connectivity, battery (power) consumption rates, different remote controls and limited input modalities give rise to uncomfortable experience for users. The more usage of devices increase, the better users require usability. The present study describes which factors and attributes are most relevant to improve usability. We place emphasis on designing methodologies of pilot study to enhance usability thorough data collection and analysis. As a result, we define usability model of multiple devices by those factors and attributes.

This study addresses the design of nonlinear oil-spring damper that render an effective re-centering mechanism for a linear motion (LM) guide base isolation system. A nonlinear stiffening behavior of the damper offers added advantage of re-centering mechanism. The proposed LM guide with damper system works in a similar fashion as that of the LRB isolation system supplemented with re-centering mechanism for small to high level of shaking. For high intensity shaking, the proposed system minimizes the peak horizontal displacement in addition to keeping the residual displacement close to zero. To demonstrate the concept of the proposed base isolation system, a numerical study is conducted with a steel moment-resisting frame when subjected to ground motions of varying hazard levels. It has been found from this study that the proposed LM guide base isolation system is effective in limiting the peak bearing displacement and making the residual bearing displacement negligible for varying hazard levels.

The study proposes the association between work zone types and work zone lane closure types used in the construction of South Korean expressways through the use of the text mining and association analysis techniques. Of the factors that greatly affect the capacity of work zones, work zone lane closure types and night work were analyzed. Words were extracted from the narrated work types using the text mining technique, and the associations among the words, lane closure types, and night work variables were identified. The analysis revealed a significant difference between the work types associated with main lane closure and shoulder lane closure. It is expected that this study can be effectively applied to establishing transportation management plans (TMPs) for work zones.

One of the most typical bridges in the Province of Daegu (Korea) is prestressed I-type girder bridges. Due to the lack of seismic detailing, this bridge may be vulnerable to earthquake events. In an effort to prevent an interruption of the transportation network, which could be catastrophic for Daegu, a comprehensive research program was conducted whose aims were to assess the seismic vulnerability of typical bridge classes. The understanding of the impact of various modeling parameters on structural-component responses is the first step in a forward-vulnerability assessment of bridges in Daegu. This paper presents part of a comparative study using detailed finite element nonlinear models that was conducted to assess the longitudinal and transverse responses of multi-span continuous prestressed I-type girder bridges in their as-built configurations using code-based loading of Korea. Deterministic responses in terms of column curvature demand, abutment footing deformations are provided for the longitudinal and transverse directions of bridge models.

Most modern office workers are seated on the chair for working with the computer during most working time. Generally, the correct sitting posture is critical for human life because wrong sitting posture can lead the back or neck disorder related sickness. The majority of previous researches have been utilized various sensors with special chairs to measure the sitting posture. However, these special chairs can be uncomfortable for users due to unfamiliar sitting environment. Therefore, we proposed the sitting posture monitoring system using Kinect which provides IR depth camera to record user’s sitting posture data and alert wrong posture to user. The proposed system provided 62% increased correct sitting posture during 35 min for 5 participants. We believe that the proposed system can help to improve the behavior of sitting posture of user without any uncomfortable adhesive sensors.

The active usage of open source software contributes many areas. However, there are many problems like ignoring license or intellectual properties infringement which can lead litigation. In this paper, we try to find original open source software by using similarity of source code. Source code similarity analyze resembles plagiarism detection problem, and using program dependence graph can be handled as a subgraph isomorphism problem which is one of NP-complete. In this paper, we apply harmony search, one of a metaheuristic algorithm, to solve the problem efficiently.

A recommendation system finds the most suitable item for customers. However, the most suitable item is sometimes refused. The most significant reason for this failure of recommendation is that we do not want what is good for us. Therefore, to make the recommendation for acceptance, persuasion should be provided. In this paper, a method of curating information for persuasion is proposed. Cognitive bias are used to determine the layout of information for customers to accept recommendations. With the proposed method, options which are beneficial but not preferable are provided without offending the customers.

An R-tree is an index structure that enables fast access to multi-dimensional data. Constructing an R-tree for a given data set yields a more efficient R-tree structure than incrementally building one as data are inserted. However it usually takes a lot of time to construct an R-tree for a huge volume of data. In this paper, we propose a parallel R-Tree construction scheme based on a Hadoop framework. The proposed scheme divides the data into partitions, builds local R-trees in parallel, and merges them to construct an R-tree that covers the whole data set. While generating the partitions, it considers the data distribution so that each partitions have nearly equal amounts of data. Therefore the proposed scheme gives an efficient index structure while reducing the construction time.

With rapid advances in wireless communications and networking technologies, the demand for high speed, wireless and ubiquitous connectivity continues to increase in order to cope with new services and applications. Increasing demand for spectral resources has introduced the problem of spectrum scarcity. Moreover, the static spectrum allocation policy for wireless communications can cause the issue of spectrum underutilization. Cognitive Radio (CR) offers promising solution for spectrum shortage and underutilization problem by means of dynamic spectrum management. Routing in Cognitive Radio Sensor Networks (CRSNs) is a very challenging task due to energy constraints, opportunistic spectrum access, dynamic topology changes as well as intermittent connectivity caused by activities of Primary Users (PUs). This paper proposes the Energy-efficient and Reliable Cognitive Ad-hoc Routing Protocol (ERCARP) with an aim to provide reliable transmission path and prolong network lifetime in CRSNs. The protocol takes account of packet loss probability, link latency and residual energy for path establishment. Furthermore, by utilizing the joint path and spectrum diversity in routing, the multi-path multi-channel routes are given for fast route recovery. The protocol performance is compared with that of the Dual Diversity Cognitive Ad-hoc Routing Protocol (D2CARP) through simulations using NS-2 simulator. The simulation results obviously prove that ERCARP outperforms D2CARP in terms of packet loss, energy efficiency, end-to-end delay and jitter.

The development of H/W and S/W has shortened the repetition cycle of new data generation and produced various categories of data. Machine learning, in particular, attracts explosive interest as it categorizes and analyzes data through artificial intelligence and contests against man. Once generated, data have their importance highlighted in terms of utilization. It is critical to analyze the data from the past and cluster new data for the utilization of data. The present study thus investigated an algorithm of determining the initial number of clusters automatically, which is part of problems with the K-means algorithm used in data clustering. The study also proposed an approach of optimizing the number of clusters through principal component analysis, a pre-processing process, with the input data for clustering. Its performance evaluation results show the accuracy rate of 87.6% or so.

A Patch Management System (PMS) distributes and manages security patches for patch-server agents after collecting the patch files from software vendors. The PMS must account for the integrity and safety of the patch files to prevent huge damage arising from possible security incidents at the agents’ environment. As software vendors cannot consider the patch compatibility of all patch-agent environments, the cause of a compatibility conflict must be analyzed when a patch fails. Existing PMSs manually verify the integrity of the patch files in a test environment. This study presents a method to automate patch testing and application, while monitoring the file modification, and reduce the time needed to analyze compatibility conflicts by using the modified file information.

In the advent of SDN paradigm, the accumulated verification technologies in the existing software fields are being used to verify the SDN. Data Plane consists of Forwarding Devices and is controlled by Control Plane. If correctness of the Forwarding Device is not verified, it affects to the whole network. However, doing every testing by manually is a huge time-cost consuming act, so it requires an automation. In this paper, it suggests a framework which applies ART (Adaptive Random Testing) technique which considers OpenFlow Switch to be Black Box from the Controller point of view and is easy to do a testing automation.

Since more systems require management by the patch management system, it is difficult for patch administrators to collect patches using existing methods. In addition, importance of managing Microsoft (MS) patch has been coming to the fore as occurrence frequency of MS related software’s vulnerability is increasing. The paper proposed a system that enabled them to automatically check and collect the MS patches and manage them efficiently through automatic patch alarm and collection using WSUS.

This paper presents a method of building a security vulnerability information collection and management system in order to promptly identify which types of security vulnerability are included in IoT devices.

Energy conservation is one of the most important issues in selection of cluster head in wireless sensor networks. Traditional cluster head selection methods reduce cluster communication distance, but there is not enough increase in the network lifetime. In this paper, we propose CHSM-RSA (Cluster Head Selection Method by Restricting Selection Area). CHSM-RSA reduces cluster communication distance by using partitioning and restricting network area where nodes can be selected as cluster head.

There is a high level of interest in pedestrian detection systems based on worldwide acknowledgement of pedestrian safety, and the need for research in this area is increasing. The HOG based pedestrian detection method proposed by Dalal and Triggs has been recognized as being less sensitive to the clothing and poses of pedestrians and also changes in lighting and is therefore one of the main methods used for pedestrian detection. But because the HOG based method requires a significant amount of computations, it is difficult to implement this method in real-time. Therefore in this research study, to improve the speed and detection rate for pedestrian detection, a pedestrian detection method that was improved upon by reducing dimensions of the particular feature vector extracted using UDP dimension reduction was proposed, and the results of performance evaluation showed that compared to previous HOG, HOG-PCA and HOG-LPP etc. algorithms, the speed and detection rate of the proposed algorithm were confirmed to have been improved.

The Internet is the global system of interconnected computer networks which use the Internet protocol suite (TCP/IP) to link billions of devices worldwide [1]. These devices are the part of Internet of Things which are used to bring simplicity to our lives like a home surveillance camera we set to our home. It is crucial to secure those connected things around us by discovering things with insecure configuration. To achieve the security measure, we are to set up a framework. Are we sure that we are the only one who can access it? Our approach is to search and gather the specific information of the IoT devices on the internet to provide a data to analyze the vulnerabilities. To search all these devices, we need to check each IP in IPv4. This task takes a long time since handshaking cannot be accomplished so fast. Our approach gives a solution to this problem.

A system that takes collected and analyzed IoT vulnerability information and transform into STIX/TAXII format to share and store, in order to effectively share information among users.

Internet-connected device information can be acquired through the open ports of a network host. It is also possible to determine whether a particular host is vulnerable by associating publicly known vulnerabilities with this information. Currently, the analysis of the device information to identify the security vulnerability is carried out manually; therefore, automatic analysis technology is necessary in order to deal with a huge number of devices. In this paper, we propose a method that automatically generates the Common Platform Enumeration (CPE) of Internet-facing devices based on banner information to discover security vulnerability information such as Common Vulnerabilities Exposures (CVE).

IoT Devices is increased rapidly, but the tool or framework for security check of IoT device is an insufficient. To improve the security of IoT device finds a weak part of source using the Source Code Auditing Tool or check the error or termination part of protocol using Protocol Fuzzer. But normal Protocol Fuzzer does not find an internal problem through fuzzing process. So we design a blackbox fuzzer combining firmware dynamic analysis platform and IoT protocol fuzzer.

Due to the recent development of wireless network technologies, the use of IoT devices is increasing as well as security threats caused by vulnerable security. To prevent security threats and accidents, it is necessary to share information on cyber security threats. This paper proposes a platform of sharing vulnerability information regarding IoT devices, and contents of shared information.

Internet of Things (IoT) becomes more popular, and things are connected to each other through wired or wireless communication methods. Though things are connected with various methods easily, it attracts network attackers who exploit these open and convenient network connections in order to obtain unjustified information and benefits or to subvert various IoT systems. Especially, Denial of Service (DoS) attack becomes a serious problem on IoT system where huge number of devices are connected to. These devices are usually connected to IoT gateways in order to send packets to Internet. However, currently the impact of DoS attack on an IoT gateway, which has various interfaces such as wireless LAN interface and wired LAN interface, is not well examined. In this paper, we assess the impact of DoS attack on an IoT gateway with various scenarios. We implemented a prototype of an IoT gateway which has wired and wireless network interfaces by using Raspberry Pi, OpenWRT, and OVS (Open vSwitch). With this prototype, we evaluated various DoS attack scenarios on this IoT gateway. Through this evaluation, we observed the severity of DoS attack on IoT gateways, especially for wireless connections.

In this paper, we developed a low-cost video-oculography device that diagnose neural diseases such as nystagmus by the vestibular function test using an infrared camera. An infrared camera and LEDs were attached in a pair of developed goggles, which was located in front of an eye for experimenters. Collected videos were converted to gray channel from RGB channel, and each pupil was extracted using morphology operation. Rotatory chair tests were conducted with our device. Gain, asymmetry and phase were calculated from obtained video.

Virtual machines are enabled to many of physical server can be integrated into fewer physical server. Integration of the server using virtual server technology induces the efficient use of resources to bring the cost benefits. Power consumption of the data center has been increased by 45% or more every year. More than 60% of the maximum power consumption is wasted on the physical server idle state, one way to reduce energy consumption is to minimize the number of physical servers. In this paper, VM usage time (running time) applied to 0–1 Knapsack algorithm. This method is the VM arrangement technology that can minimize the use of energy.

Low-Power Wide-Area Network (LPWAN) or (LPWA) is a type of wireless telecommunication network designed to allow long range communications at a low bit rate among things (connected objects), such as sensors operated on a battery. The characteristics of extremely long communication range and low power provide more feasible, reasonable wireless connectivity with a lot of researchers, field engineers, and application designers, who have suffered from weak wireless connectivity, range limitations, and energy efficiency. Therefore, in this paper, key applications and challenges of LPWA will be handled first, and several existing (emerging) new LPWA standards will be discussed. In particular, LoRaWAN, which is regarded as one of the most effective LPWA solutions, is mainly focused. So, several functionalities and characteristics of LoRaWAN will be presented. In addition, network architecture different from legacy short range wireless communications such as IEEE 802.15.4, IEEE 802.11, etc., will be overviewed and state-of-the-art off-the-shelf LoRa chipsets and modules will be introduced. Then, in order to help understand how to develop LoRaWAN application, our LoRaWAN experiments will be presented. At the end of the keynote, design considerations of LoRaWAN application and network, and several types of design methodologies will be discussed.

This study will first select the feature of consumer behavioral system and build up its model then will evaluate the goodness-of-fit to confirm its suitability through the second-order factors analysis. There are six dimensions of risk correlated with consumer BIS – Functional, Financial, Social, Physical, Psychological, Time – and four dimensions of factors correlated with consumer BAS – Quality, Efficacy, Safety, Cost effectiveness.

This paper deals with the design approaches of ICT functionality for smart city in an attempt to identify cross-conforming aspects between extreme approaches (IIFA, CSFA). Based upon a general modelling of city spaces and infrastructures, properties of those two approaches are compared to identify the chance to compensate the weak points and to combine the preferable features of those models each other. The result gives us a direction to combine those approaches for the design of ICT structure for smart city, that is synergistic enough for the planning of a smart city not only at a manageable level of fast installation but also for accommodating heterogeneity in city subsystems.

When designing Korean fonts, about 2,500 widely used characters should be designed among 11,172 characters. When generating fonts, characters are generally described as ‘outline’. On average, it takes more than 1 year to design single set of Korean font with an outline font editing systems. Also, it takes almost the same amount of time to change the style of an already generated font using general outline font editing systems. In this paper, we propose a Korean font editing system which uses font parameterization technique based on METAFONT. Korean characters are composed of basic units of strokes and radicals unlike Roman characters. They have combination rules of the basic units. Therefore, we extracted font parameters for changing font styles by considering these characteristic of Korean characters, and applied them to Korean fonts implemented with METAFONT. In addition, we developed GUI-based Korean font editing system for efficient user interaction.

The data collection of vehicle-driving GPS trajectory becomes the basis of big data analysis and prediction for a variety of purposes, such as navigation and movement analysis. In order to properly analyze a large amount of GPS location information, it is necessary to determine the exact road map and location data by matching a digital map and space. We previously discovered the road information of the GPS coordinates using the commonly utilized map-matching technique. However, such a navigation map-matching technique requires a lot of supplementary corrections in order to rapidly and accurately navigate a large amount of data. In this study, we apply geohash indexing and long link vertex dividing preprocessing to spatial data for performance improvement of massive data map matching. Also speed filtering logic is applied together for qualified analysis. We established and implemented a distributed analysis environment for the better big data map-matching with HBase. Altogether we constructed a spatial analysis system using the MapReduce mechanism, which improved its performance. This paper shows that our analysis system provides the 44 times performance achievement compared to traditional mysql DB processing with mesh structure for 5,000,000 cases of GPS trajectory.

Smart homes are increasing their popularity as the most promising application of Internet of Things (IoT). Security has becoming an important issue in smart home. There are many security threats and challenges present in smart home. To overcome these security issues, we proposed security system architecture for home automation system. The architecture is divided into three parts: management platform, secure home gateway, and home controller. The proposed system fulfills the security goals such as user and device authentication, protecting communication, and different attacks.

OpenStack cloud is an open source private cloud environment and uses the filter scheduling algorithm for scheduling of virtual machines to hosts. Filter scheduler selects the host with highest weight and assigns the virtual machine. In the process of weighing filter scheduler does not consider virtual machine weight. We propose a new scheduling algorithm for OpenStack private cloud environment by considering the virtual machine weight in the weighing process of hosts. In this paper, Round Robin, Greedy, and Filter scheduling algorithms are compared with the VM weighted filter scheduling algorithm.

To transmit and to store digital video sequences, the compression is vital. Motion Estimation (ME) is generally used to reduce redundant data in video sequences. ME which limits the performance of image quality, bitrates and encoding time require much complexity. To reduce the huge computational complexity, a hierarchical motion estimation method for multi-view video coding is proposed. The proposed method exploits the properties of motion vectors. The characteristic of the distribution of motion vectors is used to place the search points in the search area and to choose a search pattern for the current block. Experiment results show that the complexity reduction of the proposed method over PBS and TZ can be up to 98% and about 45–76% respectively while maintaining image quality and bitrates.

This paper analyzes technological evolution from viewpoint of change in technology system. As knowledge base, which used to describe technology system conventionally, suffers from heavy dependency on domain experts, this paper replaces knowledge base with hierarchical topic model to analyze the evolution process of technology system. Specifically, we find frequent closed itemsets from terminologies in patent documents at first, then discover association rules and use them to measure the importance of terminologies and semantic relationship between terminologies, afterwards we clean terminologies in corpus and run HLDA model to describe technology system, finally, we analyze technological evolution via changes of technology system. An empirical research on Hard disk drive demonstrates the feasibility of this method.

Inventor name disambiguation is a task that distinguishes each unique inventor from all other inventor records in patent database. This task is essential for processing person name queries in order to get information related to certain inventor. We proposed a mixture approach that applies to the combination of supervised learning, stochastic record linkage and ruled-based method to determine whether each pair of inventor records are from same inventor or not. Our algorithm tested on the USPTO patent database disambiguated 12 million inventor records in 7 h. Evaluation is on labeled dataset from USPTO PatentsView inventor name disambiguation competition and showed our approach have an excellent output.

A framework comprised of three types of analysis is presented for the competitive analysis on industrial technology. The three analysis are technology life cycle analysis, competitive technology analysis and competitor analysis. The first is used to comprehend current technical development stages and developing trend, and the second is used to discover developing technical topics, the development situation and relationship of sub-technologies, while the third is used to explore technical competitors, their relationship and technical strength. A case for analysis the industrial competition of fuel cell technology is introduced. The case shows the three analysis can provide valuable competitive information of industrial technology, and it convince the effectiveness of the framework.

Text classification is the task of assigning predefined classes to free-text documents, and it can provide conceptual views of document collections. The multinomial naïve Bayes (NB) classifier is one NB classifier variant, and it is often used as a baseline in text classification. However, multinomial NB classifier is not fully Bayesian. This study proposes a Bayesian version NB classifier. Finally, experimental results on 20 newsgroup show that Bayesian multinomial NB classifier with suitable Dirichlet hyper-parameters has similar performance with multinomial NB classifier.

This paper presents the application of CERIF on the electric vehicles decision support system. Firstly, it discusses why we choose the electric vehicle industry and what’s its special requirements; how do we use CERIF to respond the challenge for the data-driven decision support and how do we carry out our data management. At last, we propose a list of summary for the application of CERIF in the electric vehicle decision support activities.

The class of memory technologies with best features from memory and storage are called storage-class memory or SCM. In the hybrid usage model of SCM, it is physically attached to the memory bus just like DRAM but it is logically shown as a block device just like a storage. In this paper, we question the effectiveness of the page cache in the hybrid model, because SCM has the read and write performance comparable to DRAM. Therefore, I implemented a way to bypass the page cache in the Linux kernel, and show the effectiveness the page cache by thorough experiments with various file I/O benchmarks.

In the recent years, machine learning and deep learning has been becoming hot research titles. In many human life’s fields, AI takes big roles like auto driving a car, automatically working robots or vacuum cleaner using image recognition techniques. Tensorflow is a machine learning system with open source code was introduced and provided by Google on November 9, 2015. It has been being famously used in images recognition field. In our work, we recognize an image and classify it using tensorflow based on Convolutional Neural Networks (CNNs) and determine what it is. We train 5-layers CNNs by supervised learning from a database. After training process, trained data files are generated. In the next steps, we use this data to recognize input image and classify it. Finally, we test the results by a testing program.

Mobile devices, such as smartphones and tablet PCs, have evolved continuously from the time when they debuted in the late 1990s. At the same time, the structure and usage of mobile applications have also become increasingly complex. As a result, it is often found to be difficult to understand the user interface (UI) of applications. In addition, the low interpretability of metaphors in UIs makes the problem worse. These conditions and user environments inhibit smooth learning of applications. Accordingly, it can be inferred that the low interpretability of metaphors is expected to eventually negatively affect the learnability of applications. However, prior studies in the information systems (IS) field have not shown much interest in the effect of the interpretability of metaphors in UIs of mobile applications on the learnability of the applications. The main research goals of this study are as follows: (1) to examine the effects of the interpretability of metaphors in UIs of mobile applications on the mental model of users of the applications and on the learnability of the applications, and (2) to find the effect of the mental model of users on the learnability of the applications. The data was collected through a survey and structural equation modeling (SEM) was used for the analysis. The results showed that the interpretability of metaphors has significant effects on the mental model of users as well as on the learnability of applications.

In database outsourcing, because a service provider might be untrusted or compromised, two issues of data security emerge: data confidentiality and data integrity. Motivated by these issues, we propose a density-aware data encryption scheme and a query processing algorithm with query result auditing for secure mobile services. To guarantee the data confidentiality, our density-aware data encryption scheme utilizes a grid index to transform the original data space into a bitmap signature. In addition, to reduce the transmission overhead of verification data, we propose a query result authentication index that stores an encrypted signature for each anchor, which is the concatenated hash digest of clustered data. Through performance evaluation, we show that the proposed scheme guarantees the high level of privacy preservation to users while providing better query processing performance, compared with the state-of-the-art schemes.

Moving object segmentation is an important component of many vision systems, especially in the non-static background. This paper proposes an approach based on Kernel Density Estimation that can handle situations where the background of the scene is not completely static but contains significant stochastic motion (e.g. water). To get the initial results, a higher dimensional KDE model using the observing pixel intensity values and the information of optical flow is built. Then a KDE observing model based on the Hidden Markov Random Field Model and the Expectation Maximization frame work, is used for segmented the moving object. Experimental results show that the proposed approach can accurately detect moving objects and use less video frames.

Maximal Cliques Enumeration (MCE), as a fundamental problem, has been extensively investigated in many fields, such as social networks, and biological science and so forth. However, the existing research works usually ignore the formation principle of maximal cliques which can help us to speed up the detection of maximal cliques in a graph. This paper pioneers a novel problem on detection of bases of maximal cliques in a graph. We propose a formal concept analysis based approach for detecting the bases of maximal cliques and detection theorem. It is believed that our work can provide a new research solution and direction for future topological structure analysis in various complex networking systems.

The study designs and implements a database system for predicting small region-specific cultivable crops based on future climate change utilizing integrated public Big Data. For this study, regional temperature factors, regional precipitation factors, land acidity, solar radiation, cloud amount, and appropriate climatic factors for each crop were utilized. The database system could extract the information of each small region such as kinds of currently cultivating crop, kinds of regional cultivable food crop, kinds of regional cultivable fruit, kinds of regional cultivable medicinal crop, kinds of regional cultivable vegetable, and changing trends of each crop production quantity. Based on these small region-specific crop information, it is possible for the farmers to increase future profits of farm households by providing information of medicinal crops, food crops, vegetables, and fruits that can be produced in each regional farmhouse. It is also possible to present future recommended crops to individual business operators by utilizing these public big data, to suggest the need for development and research on crops that can be cultivated in each region, and to suggest marketing plans for present and future crops.

This research statistically analyzes factors that influence consumer intention to use virtual reality services. The result shows that the main predictors of intention to use virtual reality services, in the order of importance, are hedonic motivation, personal innovativeness, effort expectancy and performance expectancy. And it shows that the higher were the impacts of effort expectancy, social influence, performance expectancy, and hedonic motivation on intention to use the services the higher was a customer’s personal innovativeness. According to the results, marketing strategies for virtual reality services should appeal to consumers by positioning the using experience as an adventure or a way to reduce their stress and change a negative mood. Also they should be reputation-building and target early adopters.

This study has the goal of developing a diagnosis system to detect human depression in real time to assist in the diagnosis of a doctor. The developed system may grasp the concentration and depression level of a patient using brainwave data acquired in real time. The depression detection index used in the system is the frontal brain asymmetry (FBA), which is based on the asymmetric phenomenon of depressed patients. In this study, an experiment was conducted with 40 depressed/normal subjects in order to verify the reliability of the developed system. The results proved that the system diagnosed the depression level in real time. It can be used to develop therapy programs for various nervous and mental disorders.

Face micro-expression is crucial for feeling perception and yet demanding due to the high dimension nature and the increasingly request for the recognition accuracy. The tradeoff between accuracy and efficiency by Deep Belief Network is a challenging. This paper shows that a two-stage strategy can achieve both speedup and high accuracy. With it, an efficient facial micro-expression algorithm is proposed that consists of Double Weber Local Descriptor devised in this paper firstly for extracting initial texture local features, and Deep Belief Net for more global feature and less computation dimension. The experiments with JAFFE database show that the average recognition rate by the new algorithm is up to 92.66%, and the rate of neutral facial expression is nearly 100%. Compared with LBP, LDP, PCA, Gabor wavelet and Weber local descriptor combined with DBN, the new algorithm of the introduction of Double Weber Local Descriptor into DBN has higher recognition rate.

Network forensics is a comparatively new field of forensics science. The growing popularity of the Internet means that computing has become network-centric and data is now available outside of disk-based digital evidence. To collect certain network data for forensics, real-time network file packet inspection becomes a hot topic as it is needed in many applications such as virus detection, intrusion and attack forensics. Most of the traditional techniques use exact matches on keyword and/or white/black MD5 lists to have an efficient inspection. However, it is well-known that exact matches may not be effective to identify similar files such as the same videos with small changes, e.g. titles, posted by different users or metamorphic viruses (mutated computer viruses). Approximate matching is known to be more robust to identify similar files and has been proven to be effective in digital forensics. In this paper, we design a network forensics system by recording objective network files for future analysis. We try to confirm that by using an appropriate approximate matching approach, it is feasible and effective to inspect real-time traffic in order to identify similar files. Our experiments with real data show that our solution achieves good usability in practical.

With the active sequencing studies, along with the advancement of the next-generation sequencing (NGS) technology, rapid progress has been made in genome analysis of various species. For the completion and analysis of the whole genome map, it is necessary to assemble the read results from NGS data for map completion. When reference models are available for assembly, similar sequences can be used for mapping assembly, whereas the de novo assembly method is applied when there are no models available. At this time, if the number of repeats in repeat regions is unclear, it would be challenging to assemble the whole genome map. Thus, the aim of this study was to conduct comparative analyses of the repeat regions using assemblies from various assembler tools automatically yielding repeat regions, and to carry out effective assembly analysis including repeat regions.

Current demographic trends show that there is a growing number of older population groups, which might result in serious social and economic problems in future. Therefore national governments want to prevent them by implementing different strategies which could contribute to the maintenance and enhancement of quality of life of older people. One of the approaches to handle this issue is also traveling which can be enhanced by information and communication technologies (ICT). The purpose of this article is to explore the use of ICT by senior travelers. This is done by literature search of available studies on the research topic in the world’s databases Web of Science, Scopus, ScienceDirect, and Springer. The findings show that senior tourists are now more technologically savvy than they used to be two decades ago. In addition, their use of ICT reflect their confidence and independence in travelling. Nevertheless, since senior tourism is a new developing branch of tourism, much more research has to be done in this field, including the research on the use of ICT by these older travellers.

Currently, there is a growing number of older people worldwide. This demographic change results in serious social and economic problems. Therefore governments, especially in developed countries, attempt to intervene in this process and help to enhance quality of life of older population groups with different means. One of the approaches in this respect is the use of mobile and wireless technologies in healthcare. The purpose of this article is to discuss the use of wearable and portable monitoring devices for older people in three areas of healthcare: fall detection, dementia care and low access to healthcare. The findings revealed that there is a lack of clinical studies examining the use of wearable devices in healthcare for older people. Therefore more research should be performed because the benefits of the use of wearable devices, such as their unobtrusiveness, sensitivity, or reliability, can contribute to the enhancement of quality of life of older people.

Anomaly detection is very crucial in an intrusion detection task since it has capability to discover new types of attacks. The major challenges of anomaly detection are how to maximize the accuracy while maintaining low positive rate. In this paper, we propose new approach on anomaly detection using multi-level classifier ensembles. We employ an ensemble learner as a base classifier of ensemble rather than a single classifier algorithm. We run several experiments to choose the best combination of two-level classifier ensemble model. From our experimental result, it is revealed that the performance of our proposed approach yields satisfactory results over classical classifier ensembles and single classifiers.

As we use smartphones as daily use as televisions and cars, smartphones are always used for daily life. And information technology and sensor technology matures enough to detect user’s movement, direction, shapes and bio-signals. With sensor technology and mobile learning device (smartphone, tablet PC) with sensors, it is possible to support and satisfy instant personal learning activity tracking and learner’s learning emotional state for learning environment and learning contents. Lastly, on distance learning environment or e-learning environment, learning interactions and feedbacks between learners and teachers is almost impossible, since there is online communication and learning contents which can not deliver or transmit learning emotion of learners. But, in order to deliver personalized learning contents and modify learning environment according to a learner, learning emotional state from learning contents is very important. Learning environment state means analyzes learning environment information consists of learner’s property, learner’s environment and learner’s activity from learner’s e-portfolio and learner’s educational devices. And besides learning environment states, real-time learning emotional state (learning emotion state) from learning contents is necessary.In this paper, we design Learning Reaction Analysis Engine for interactive digital textbook platform. Proposed interactive digital textbook platform has Learning Reaction Analysis Engine that could detect learner’s learning emotional state that means emotional learning interests and learning concentration state, and Learning Reaction Analysis Engine can analyze learner’s learning emotional state from learning activity state, learning device execution state and learners’ bio-signals from smartphone, tablet PC, or smart watch. In order to analyze learning emotional state information, we construct learning emotional model that defines learning emotional information, and classification and inference rules of learning emotional information. Learning emotional state inference rules can be applied into Automated Tutoring Engine and Personalized Learning Contents Modification Engine. Learning Reaction Analysis Engine decides learning strategy for type of learning contents frameworks, learning sequence, difficulty change of learning contents and learning contents modification for current learning contents according to a learner who is accessing to the learning contents.

Many users in social web environments share and publish user-generated contents such as tastes, opinions, and ideas in the form of text and multimedia data. Various research studies have been conducted on the analysis of such social data, which can be used for discovering users’ thoughts on specific topics. But, there are still challenging tasks to find out the meaningful patterns from the social data due to rapidly increasing amount of data. In this paper, we therefore propose a rule-based topic trend analysis by using On-Line-Analytical Processing (OLAP) and Association Rule Mining (ARM) to detect information such as previously unknown or abnormal events or situations. For the verification of the proposed method, we conduct experiments to demonstrate that the method is feasible to perform rule-based topic trend analysis.

In this paper, we present the hierarchical safety analysis for eliciting traceable safety requirements. The proposed technique was used to the case study of railway system as an example. In this work, FMEA and HAZOP analysis are used as safety analysis technique in order to illustrate hierarchical safety analysis showing traceability.

With the rapid expansion of new technologies in all spheres of human life, technological devices have inevitably penetrated in people’s homes, where in most cases are targeted at facilitating people’s life in order to improve people’s quality of living. The purpose of this article is to address current issues with respect to smart home technology and energy savings, especially in European countries. The specification is based on the basis of available studies between 2010 and 2015. The authors used a method of literature review of available sources exploring research studies focused on smart home technology with respect to energy savings in the acknowledged databases Web of Science, Elsevier, Science Direct, and Springer. The findings show that the number of studies exploring the issue of smart home technology and energy savings is gradually rising since consumers are getting more aware of potential savings, respectively of waste of energy.

The aim of this paper is to specify the seniors’ attitudes towards modern technology as a tool for improving the quality of life in case of chronic disorders. The research is conducted in the Czech Republic. This study employs the analysis of both primary and secondary data. Primary data was collected through a questionnaire survey among senior citizens living in the Czech Republic. 170 questionnaires were distributed, 112 of which were returned and processed. The source of secondary data was the Czech Statistical Office. The case study done in the Czech Republic showed that local patients’ attitude to modern technologies is rather positive. 86% of respondents use personal computers and 32% wield smart phones. Patients tend to be hesitant only concerning the utilization of smart household appliances, which would be welcome by 44% of the respondents and only 26% of them agreed strongly.

Human counting in cinema is easily influenced by varied illumination, so as to become a complicated problem. This paper develops an audience counting system in cinema by detecting occupied chairs in captured images. Firstly, we initialize chair regions in a background image manually. Then, the differences between the background and current images are detected as foreground regions. Such rough segmentation results always contain noise because of environmental illumination changing. Thus, a contour difference detection algorithm is applied to refine the audience detection results. Next, if both foreground and contour differences in a chair region are larger than a threshold, this chair is recognized to be occupied by an audience. Finally, the audience number is estimated by counting the occupied chairs.

To the best of our knowledge, there has been no research on index structure for the encrypted data. In addition, the existing query processing schemes over the encrypted data can support limited types of queries [1–3]. To solve the problems, in this paper, we propose a distributed index structure and a query processing scheme for the encrypted data. The proposed distributed index structure guarantees data privacy preservation and performance improvement for the various types of queries. In addition, the proposed query processing scheme provide both high query performance and 100% accuracy while preserving the data privacy. Finally, we show from our performance analysis that our proposed index structure and query processing scheme are suitable for protecting the data privacy of the mobile sensitive data.

Due to advancement in cloud computing technology, an order-preserving encryption schemes, called Programmable Order-Preserving Secure Index (POPIS), has been proposed. This scheme hides the original data while keeping the order of the encrypted values the same as that of the original data. So the service provider can perform query processing without decryption. However, because the encrypted data in POPIS is sorted by certain column values, it is weak to both order matching attacks and count attacks. To solve this problem, we propose a data encryption scheme using periodic functions. Our scheme generates encryption signatures based on data groups and periodic functions. With this, we can preserve the order of each data group and also can guarantee the data privacy. Finally, we show from the performance analysis that the proposed scheme is better in terms of the degree of privacy protection than the existing data encryption scheme.

In this paper, we describe the software product line process centered for small and first introduced company. Most software product line process focused on domain engineering and application engineering since variability management and product configuration technique are main differences from other software development methodologies. When the organization considers the adoption of new methodology, manager consider the whole lifecycle management. In this paper, we define whole lifecycle for software product line engineering and explain the experiences of real pilot project. Our suggested model can be used as software product line engineering transfer model.

In recent years, flying Ad-hoc networks (FANETs), which consist of small unmanned aerial vehicles (UAVs), is being used in the increasing of civilian and military applications. Due to the high mobility of the UAVs nodes, the link between the UAVs may frequently be disrupted. Hence, the existing routing protocols are inability to perform in FANETs. Motivated by this, we propose a new routing protocol named UAV-OLSR for FANETs in this paper. This protocol is based on the well-known protocol called optimized link state routing protocol (OLSR). We focus in our protocol on the lifetime of a communication link between the UAVs nodes and named link live time (LLT). We propose a new multipoint relay (MPR) selection algorithm where the UAVs node with maximum LLT is selected as the MPR. Our emulation results show that UAV-OLSR protocol outperforms OLSR in the packet loss rate, total time delay, the average time delay, and traffic received.

Similarity join is an important operation in MapReduce framework to find pairs of similar objects like images, video and time series. Since MapReduce basics do not support efficient join processing, the duplicate reduction of candidates and load-balancing among partitions are the major challenges. Recently, many partition based similarity join algorithms have been proposed to solve such problems. However, the existing algorithms still have limitations for supporting efficient join processing over large-scale data set. In this paper, we proposed a similarity join algorithm with an efficient filtering technique on MapReduce to overcome the limitations of traditional partitioning method in two ways: (1) the number of outputs records generated by the filtering matrix reduces duplicates and (2) the estimated join cost generated by using a partition matrix leads to a better load-balance among reducers. Moreover, we have conducted experimental evaluations using sequential data to show the speed-up and scale-up of proposed method.

Recently the use of computing technologies for distribution systems such as cloud or grid is increasing for effective management of IT operation and investment cost reduction. One of the important tasks in distributed systems is to prevent the server from overloading. Although a load balancer has been introduced to solve this problem, there is a limitation that an existing load balancer must use only one initially set algorithm. To solve these limitations, this paper proposes a load balancing method that dynamically replaces the load balancer algorithm according to the server status and environment. The proposed method is that the load balancer collects the server status information in real time, analyzes the result of the server status, and replaces it with an appropriate algorithm if necessary. We experimented with several algorithms and proved that this proposed load balancing is more effective than the existing load balancing.

Face image with illumination variation usually contains redundant data that will seriously reduce the recognition rate. To combat the influence of illumination variation and extracting illumination-robust feature, a novel feature extraction method is proposed. The novel method is based on the combination of Center-Symmetric Local Binary Pattern (CS-LBP) and the fusion of the vertical and horizontal component images derived from wavelet decomposition. Numerous experiments have been done on the Extended Yale B to verify its effectiveness. The experimental results show that by applying the proposed method, redundant data caused by severe illumination variation can be filtered, while useful texture information can be reserved and enhanced. Compared with CSLBP, it significantly improves the face recognition performance under severe illumination variation.

Formal analysis of cryptographic protocols is to find out flaws in the protocols by various formal methods. Some supporting tools for formal analysis of cryptographic protocols have been proposed and applied, but the tools failed to support the whole processes of formal analysis automatically. Therefore, a supporting environment which can support formal analysis automatically is needed for analysts. This paper presents the first supporting environment for formal analysis of cryptographic protocols.

LPWA networks are getting attention as a solution to support massive number of IoT devices. LoRa is one such low-power based long-range technology. In this paper, we discuss methods to build a LoRa simulation environment. We first discuss the characteristics of the LoRa PHY and MAC layers, defined in the LoRa and LoRaWAN specifications. Then, we show how LoRa PHY and MAC functions can be realized in the OPNET simulation environment. For LoRa PHY implementation, we adopted LoRa modulation curve based on BER vs Eb/No. We also implemented various process models depends on LoRa node model for LoRa MAC functions. We conclude with future directions for performance enhancement.

Malware attacks have been posing various security threats such as data losses, personal information and financial information, system damage, and IT infrastructure destruction. To prevent these security threats in advance, many anti-malware programmers and malware analyzers have been analyzing malware. But methods of attacks are diversifying and it makes it harder for analyzer to analyze malware. For instance, bypass dynamic analysis malwares such as time-trigger are much more difficult to analyze than general malware because its function is executed at a particular time. In this paper, we proposed that automatic analysis of bypass dynamic analysis malware such as time-trigger. First, for our proposal, we utilizes BFS (Breadth-First Search) algorithm to track malicious behaviors flows from the beginning to the end. And such flows of malicious behaviors were visualized into graph. Furthermore, we calculated malware similarity using SSIM (Structural Similarity Image Metric) based on malicious graph.

Web advertising is being given various opportunities to individuals and business enterprises. Web advertising has following advantages such as distributed directly announcing or data and marketing events without barriers of time and of location to the customers.Aim of this study is to emphasis the facing problems, interrelation and attitudes of internet customers of Mongolia at websites of advertisements. This study based on survey of customers and 500 customers of internet participated at this survey. As a result of this study, information feature of web advertising and customers’ attitudes at website of advertisement are positive and impact the deepest influence. As a studying their real consumption and customers’ attitudes at web advertising, communication of customers and service providers will be improve and can exchange safety information each other.

The serious game which is utilized for educational training has been operated using I/O (Input/Output) devices like keyboard, mouse and monitor. Recently, serious game is advancing towards VR (Virtual Reality) based training system by providing interaction functions between virtual visualization environment and real action of trainee by wearable devices, such as motion recognition system and HMD (Head Mounted Display). For acceptance test of VR based training system, hardware testing is possible by using test metrics which can be described on its specification. But VR based training contents as a software including virtual visualization environment has many difficulties for testing due to lack of quantitative test metrics about realism which means similarity between real environment and its virtualized environment. In this study, it is suggested quantitative design and documentation methodology of VR based training contents using test models derived from ISO/IEC 25010:2001 SQuaRE (System and software Quality Requirements and Evaluation). The suggested quantitative design document has been experimented through the development of a VR based plant safety training system and has verified its usefulness as training contents test metrics for testing realism which is an emotional factor.

This paper proposes a motion artifact (MA) removal method in the photoplethysmographic (PPG) signal for accurate heart rate estimation. PPG signal is easy to acquire, but it is easily distorted by body movement. In this study, MA is analyzed using acceleration signals and removed in the PPG spectrum for accurate heart rate estimation. The proposed method progressively removes three-axis acceleration spectra in order of spectral power. The performance was confirmed by comparing heart rate estimation errors one case that MA was removed with another case that MA was not removed. After removing MA and applying two peak tracking methods in 12 data sets, the mean absolute error (MAE) of the beat per minute (BPM) is lower than conventional methods.

The purpose of this paper is to confirm the improvement of accuracy in predicting the profit of a café by using dimensionality reduction features through Factor Analysis. Profit forecasts for retailers have always been of great interest. We limit the discussion to the prediction of a café profit. We show that dimensional reduction through Factor Analysis is useful for various types of data. After that, we compare the SVM with the linear regression and show that using a good kernel trick of the SVM improves accuracy.

In this paper, we propose a new algorithm that renders motion blur and motion blurred shadows at the same time using a hybrid approach. Our algorithm generates a shadow map which stores a list of visible time ranges along with depth values at each pixel. In the subsequent pass, we use this shadow map to perform shadow tests at a receiver sample’s position and at its time. Our results show that our algorithm addresses some problems that a previous work does not. In addition, our algorithm runs completely on the current GPUs.

Thanks to the design of different portable subminiature sensors and wired and wireless communication technology, the U-Healthcare service is getting vitalized. A mass amount of raw data is processed in real time when this U-Healthcare service is provided, and efficient processing and storage technologies are required accordingly. Therefore, this paper proposed an ECG data compression algorithm that is improved to efficiently transmit M2M-based mass biometric data.

Recently, there has been an active research effort on Wireless Sensor Network (WSN) where the sensor nodes consume energy efficiently by communicating between the nodes directly without a network infrastructure. However, previously proposed protocols require regular re-establishment of clusters, which leads to unnecessary energy consumption. Moreover, there is a large energy consumption because a cluster head that is placed far apart from a sink node directly transmits data to the sink nodes. Therefore, in this paper, we analysis the problems of the previous clustering techniques and protocols, and designed a clustering algorithm for efficient energy consumption through the use of an energy threshold during cluster re-establishment and data transmission route selection.

Cars are increasingly being equipped with a variety of driver convenience and safety features, with many vehicles evolving into what are now called “smart” cars. The convenience and safety systems built into these vehicles, which include infotainment, connected car, and autonomous navigation systems, are being actively developed by combining them with mobile communication technology. However, utilizing such technology can potentially leak personal, or vehicle information. A mobile communication module for example, can be used to attack the electronic control device of a vehicle; the attacker then has the ability to endanger the driver by gaining control of the vehicle brakes and steering devices. To protect the driver and the vehicle from such risks, we have designed a technology that creates a secure zone for the storage of important information, restricting external access to the telematics control unit.

Intrusion detection provides important protection for network security and anomaly detection as a type of intrusion detection, can recognize the pattern of normal behaviors and label the behaviors which departure from normal pattern as abnormal behaviors. We think that the traditional methods based on dataset do not satisfy the needs of dynamic network environment. The network data stream is temporal and cannot be treated as static dataset. The concept and distribution of data objects is variety in different time stamps and the changing is unpredictable. Therefore, we propose an improved data stream clustering algorithm and design the frame of anomaly detection according to the improved algorithm. It can modify the established model with the changing of data stream and detect abnormal behaviors in time.

With the continuous development of computer network technology, traditional intrusion detection system is short of good adaptability. Aiming at the traditional intrusion detection system is difficult to adapt to the increasing amount of data demand for real-time processing capability, this paper proposes a clustering algorithm based on sliding window data streams, based on which we build the IDS network security defense model. The experiment results show that the model is able to adapt to the high-speed network intrusion detection requirements.

A short text classification method based on combination of KNN and hierarchical SVM is proposed. First, the KNN algorithm is improved to get the K nearest neighbor class labels quickly, so as to effectively filter the candidate classes of documents. And then classify them from top to bottom using a multi-class sparse hierarchical SVM classifier. By this way, the document can be classified efficiently.

This letter introduces a total variation (TV) image restoration method with adaptive Regularization parameter. Our main contributions are two folds: (1) a novel selection scheme that determines the Regularization parameter of TV model in a global way through exploiting the concept of TV spectral response; (2) an efficient algorithm integrating an estimation-and–renewal strategy and the alternating minimization numerical technique to fast calculate the model solution. Experimental results on degraded images indicate the improved performance of our method, both in visual effects and in quantitative evaluations.

Face recognition system encounters complex change that varies over time, due to a limited control over the environment. So, the facial model of an individual tends to diverse from underlying distribution that collected during initial enrollment. However, new samples that are obtained each time people try to recognize or authenticate can be used to update and refine the models. In this paper, an efficient semi-supervised learning strategy is proposed to update the face recognition model. To maintain a high performance, we exploit a probability based update approach. Performance is assessed in terms of accuracy and equal error rate (EER). Experimental results illustrate that the proposed method effectively update the classifiers.

Identifying a person across cameras in disjoint views at different time and location has important applications in visual surveillance. However, it is difficult to apply existing methods to the development of large-scale person identification systems in practice due to underlying limitations such as high model complexity and batch learning with the labeled training data. In this paper, we propose a prototype system design for large-scale person re-identification that consists of two phases. In order to provide scalability and response within an acceptable time, and handle unlabeled data, we employ an agglomerative hierarchical clustering with simple matching and compact deep neural network for feature extraction.

There is increasing interest in text analysis based on unstructured data such as articles and comments, questions and answers. This is because they can be used to identify, evaluate, predict, and recommend features from unstructured text data, which is the opinion of people. The same holds true for TEL, where the MOOC service has evolved to automate debating, questioning and answering services based on the teaching-learning support system in order to generate question topics and to automatically classify the topics relevant to new questions based on question and answer data accumulated in the system. To that end, the present study proposes an LDA-based topic modeling. The proposed method enables the generation of a dictionary of question topics and the automatic classification of topics relevant to new questions.

The data centers contribute to high operational costs and electrical energy will be consumed in enormous amounts. One of the most complex challenges of energy consumption is power management. Many different methods have been applied in order to reduce energy consumption. In this paper, we propose the architecture framework focuses on analyzing the EAP (Energy-Awareness Predictor) to improve the energy efficiency. Through analysis and various integrated sensor devices, the EAP architecture framework can understanding of the consumption patterns and can better controlling of the major energy consuming. Based on inputs independent variables (value of external and internal environmental) is prediction and implement refrigeration and process control, optimization and energy management.

The structure tensor known as a second moment matrix which integrates the local data information of the image. It has been a well-established tool in the image processing field. To date a variety of nonlinear structure tensors have been emerged. Among them, the non-local structure tensors (NLSTs) is the focus of researching for the reason that it explores the spatial interactions in images. However, the performance of the existing NLST in image analysis is limited. In this paper, we propose a new structure tensor calculation method by using the nonlocal means filter to smooth the matrix-valued data. The resulting nonlocal structure tensor is effective in the orientation estimation and structural analysis of the image. Meanwhile, the nonlocal TV model based this structure tensor has been successfully applied in noise removal. Experimental results show that our model has better performance in preserving the structures, details and textures.

Recently, the image denoising methods based on patch priors have received extensive attention. Among these methods, expected patch log likelihood (EPLL) has achieved great success, using Gaussian mixture priors by the Gaussian mixture model (GMM). In the paper, we observe that GMM model requires the estimation of a global parameter $$ \uplambda $$, rather than locally adaptive parameters. Based on this, we propose a modification of the GMM model which is imposed the local constraints on partition of the image. The experimental results illustrate that our proposed method performs comparatively well.

As mobile communications develop, automobiles are becoming increasingly computerized. Connected cars can communicate in various ways, with the range of communication also expanding. Onboard communication enables the exchange of information between various control devices, including instruments in the car, engine control units, and steering and braking devices. External communication can inform the driver of traffic-related, or accident-prevention information. By connecting externally using telematics, new services including software updates, remote diagnosis, emergency calls (eCall), payment, internet, infotainment, and automobile related apps have become available. Whilst connected cars are evolving into communication devices with many capabilities, in the future connected cars may include many interfaces and gateways, in which internal and external operations can be manipulated. Therefore, a comprehensive security architecture is required within the vehicle at points of contact with the outside world, and security enhancements are required for secure storage of keys used for authentication and communication security. An approach that takes into account both functional safety and information security has thus far not been found; there is therefore a need for specialized hardware and software to protect the relevant data, along with the solutions listed, to meet the functional safety required by ISO 26262.

UNIdirectional security gateWAY (UNIWAY) is a technology that makes it possible to communicate only in one direction physically, unlike a general network that communicates bi-directionally. It is a technology to transmit data from the safe area to the non-safety area. Since it is an industrial system that operates continuously for a long time, it is necessary to grasp the normal functional operation of the system and satisfy the performance requirement in advance. In order to test that UNIWAY software operates stably for various network situations, it is necessary to elicit test items such as communication node, transmitted data size, transmission data type, and prepare test data in various combinations. In addition, since it takes a lot of time and effort to check the stability, it is necessary to automate the test and confirm the test results. In this paper, we design an automated agent that can test automatically, and repetitively using predefined test scripts and data.

Multi-step prediction for time series is a challenging research area with broad applications which can provide important information for relevant decision-makers. Many works extended different architecture of artificial neural networks to perform time series prediction, but they mostly only consider the time series itself, does not weigh the impact of time series of relevant factors. In this paper, a new method of time series prediction based on factor mining is proposed. By analyzing target time series, the means of discovering factors influencing time series and pinned down the most relevant factors was proposed. In the end, a method to do multi-step prediction with artificial neural networks, MTPF is proposed to conduct the time series prediction, create time series model and forecast time series. The proposed method is applied for a shipping price index time series prediction. Results show that this method can improve accuracy of prediction when compared with traditional methods.

The World Wide Web is a large source of information that contains data in either Surface Web or Deep Web. Compared with the data in the Surface Web, the Deep Web contains a greater amount of structured data with higher quality, but it is difficult to use directly. Studies in this field have revealed some methods for Deep Web Form Extraction, they may fall into the following categories which are HTML-based, vision-based, ontology-based, ML-based, NLP-based and so on. This paper try to combine the DOM tree and the convolutional neural network together and then find out the form in the Web page. This paper proposed a vision-based method VBF, which figures out the form from the Web page through the acquisition of the HTML code and screenshots of Web pages, establishment of the DOM tree and the calculation of the neural network and form recognition, matching, and generation.

Due to the development of deep learning, word embedding has been introduced into nature language process. So, we tried using word embedding to simplify the information extraction of questions classification and take the advantage of big data. Additionally, with the advantages of attention machine, RNN in machine translation could consider the middle states to avoid the problem of bias on inputs. We introduce it into questions classification, and the experiment shows we get a little better performance than the best performance before.

A connection between an industrial control network and IT network can expose measurement equipment, control systems and important infrastructure components to various cyber-attacks. Many technologies have been proposed to protect industrial control networks against cyber-attacks and to provide confidentiality, integrity, and availability. Among the technologies, a physical unidirectional security gateway provides protection of critical systems by forcing unidirectional communication between the two networks. The unidirectional security gateway needs to provide safety and reliability, and to guarantee, the common criteria for information technology security evaluation is operated. In this paper, we propose a unidirectional security gateway satisfying security functional requirements derived from CC v3.1.

In this paper we propose to develop technology for recognizing Jersey number of each athlete in real time to do sports video indexing in multi-player game. More than two persons are often detected as one object due to frequent collisions between persons in sports. This system recognizes the jersey number of more than one persons based on learnt context information using texts and uniform color information. The resulting system is able to achieve a jersey number recognition accuracy up to 97% on video and photography.

A variety of information security solutions are being developed to enhance user security from advanced and enhanced security threats. Highlight the aspects of end-user security as the most important aspect of information security. Based on the factors influencing the user’s behavior, this study focuses on the impact of information on the impact of information on the time required by users (Higgins 1996). The purpose of this study is to understand the causes of information security compliance and establish effective security systems to address information security policies in the information security section of the Information Security Administration. In particular, it seeks to analyze the security compliance issues and measure the outcome of the survey by focusing on visual factors and opportunities in terms of users’ voluntary improvement. To do this, consider influencing factors that affect user decision making based on self-determination theory (SDT).

An accident occurs if you are driving drowsy on a highway or if you are not looking ahead in the school zone. In this paper, we propose ‘safe driving aid system. To do this, two cameras are used. One camera recognizes the driver’s face and eyes, and the other camera recognizes the Pedestrian in front of the driver. This makes it possible to prevent pedestrian accidents and sleepy driving accidents in advance.

The Internet is one of important thing in an individual’s life and it has made a big changes. As many activities are performed by the Internet, we need specific system to prevent malicious Internet bot programs that take advantage of this convenience. Among them, security technology applying ‘Turing Test’ to distinguish whether a subject using a specific service is a human or machine is an important thing of security technology of computer science. A representative example is the CAPTCHA. However, the vulnerability was revealed by various studies and cases. For this reason, we propose a new user authentication method using sound and image.