Earticle

The rise of smartphones and computer tablets has been shifting computer tasks to mobile applications. Educational environments are changing quickly as a generation of learners grows up with technologies such as tablets, smartphones and other smart devices. With the success of mobile textbook market, tablets are replacing paper textbooks to digital. Especially, one of the most important features of electronic textbooks application is creating and reading annotations. Most electronic book readers support this fundamental functionality. However, when an application moves from desktop pc to mobile device, its feature should be redefined to fit on a mobile device. Therefore, we propose a user interface which supports visualized annotation on touch-based tablets. It enables a user to gain additional information from annotation style and guide to navigate on useful resources. This interface is developed on a tablet device to evaluate it in a classroom study to prove its effectiveness. As a result, the proposing approach encouraged students to use the electronic textbooks and annotation feature in an efficient way.

Access control is an important security mechanism that is used to limit user access to information and resources and to prevent malicious users from making unauthorized access. Traditional access control models are well suited for centralized and relatively static environments in which information about subjects and objects are known in priori, but they can hardly meet the needs of open and dynamic network environments. Access control in open network environments must therefore adapt to dynamic addition and deletion of subjects and objects. In this paper, we use game theory to analyze trust-based access control to help compute trust values that involve several factors. By viewing access control as a game played between the requester and the provider entities, we can develop strategies that would motivate subjects to make honest access to objects continuously to get the most payoffs.

Eye detection and tracking can be used in intelligent human-computer interfaces, driver drowsiness detection, security, and biology systems. In this paper a new method for eye detection based on some new rectangle features is proposed, with these features the Adaboost cascade classifiers are trained for eye detection. Then with the characteristics of symmetry of the eyes some of the geometric characteristics are adopted for correction. The geometric characteristics improve the accuracy of the eye detection, and make the rough cascade classifier trained by few samples become a reality in application. In this paper, we present an integrated eye tracker to overcome the effect of eye closure and external illumination by combining Kalman filter with Mean Shift algorithm. Results from an extensive experiment show a significant improvement of our technique over existing eye tracking techniques.

The depiction of the irregularity in nature is crucial for the realistic representation of the virtual world. For example, real pavement blocks are not always paved evenly, but partially sink over time. Many games or virtual reality systems however represent the pavement too neatly with perfectly flattened surfaces and clean texture images. In this paper, we suggest a method for representing realistic pavement blocks by adding irregularity to blocks. We create uneven surfaces by placing pavement blocks on the randomly generated terrain. We also represent worn surfaces by adding crack patterns to the original texture images along the user-controlled spline curves with Poisson equations. Our method can generate diverse new crack patterns from images of sample patterns and it would be useful to generate random patterns of other linearly-shaped textures as well. Our method is applicable to real-time rendering systems such as games and virtual reality systems since the cracking effects are incorporated into the texture images.

Breast cancer is reported as the second most deadly cancer in the world and the main of mortality among the women, on which public awareness has been increasing during the last few decades. This is why several works are made to develop help tools for disease diagnosis. Computer-Assisted Diagnosis (CAD) is based on 3 main steps: segmentation, feature extraction and classification in order to generate a final decision. Classification phase is the key step in this process; for that, many research have been accentuated in this domain and many techniques were be proposed. Kernel combination is a current active topic in the field of machine learning. It takes benefit of classifier algorithms. it allows to choose the kernel functions according to the features vectors. The combination of Kernel-based classifiers was proposed as a research way allowing reliability recognition by using the complementarily which can exist between classifiers. This study investigated a computer-aided diagnosis system for breast cancer by developing a novel classifier fusion scheme based on fusion of three support vector machine classifier. Each one is associated with an homogenous family of features (Hu moments; central moments, Haralick moment) as efficient learning algorithm and diversity between features family as fusion criteria to ensure best performance. Our experiments demonstrated that developed system using Database for Screening Mammography (DDSM) database achieve very encouraging results when compared with past works using the same information.

Recently, the cooperative communication has been proposed and studied to improve the system performance in wireless communication environments. In this paper, we propose the STBC (Space Time Block Coding) cooperative communication system with interference cancellation scheme. The multiple cooperated relay nodes that help the source signal forward are considered to be the multiple transmit antennas. The STBC scheme is applied to achieve higher diversity gain without the transmitter knowing the CSI (Channel State Information) and to decrease the complexity of the transceiver. However, if errors occur at relay nodes because of the frequency selective fading channel, error propagation would lead to degradation of the system performance when the decoded signal is retransmitted to the destination. To solve this problem, interference canceller is employed at the relay nodes when we decoded the source signals in the proposed system. Simulation result shows that better performance is achievable when interference canceller and STBC schemes are used in the PB/MC-CDMA (Partial Block Multi-Carrier CDMA) system.

Nowadays, avatars are widely used in games and Internet environments. Especially, video game consoles such as Wii (Nintendo) use avatars for representing the user's alter ego. There are several ways to generate avatars. Most existing games or Internet services provide manual systems for generating avatars. Many researchers have suggested automatic avatar generation methods, most of which generate avatars by simplifying images using non-photorealistic rendering techniques. In this research, we suggest an example-based method that generates avatars by first matching the most similar avatar components for each facial feature and compositing them. We built a system that generates an avatar which is similar to the input front-view photograph by matching each facial feature to the corresponding feature of the avatar. The system first extracts six facial features from the input image: 2 eyes, 2 eyebrows, a mouth, and a face outline. Then it matches them to the corresponding avatar components using graph similarity and Hausdorff distance. Finally, the system generates an avatar by compositing the most similar components. We have also experimented on the effectiveness of our approach, and the results show that our system generates avatars which successfully represent their corresponding photographs.

In many applications such as sensor networks, similarity search is more practical than exact match in stream processing, where both the queries and the data items are always change over time. The volumes of multi-streams could be very large, since new items are continuously appended. The main idea is to build a small size of synopsis instead of keeping original streams by using our proposed techniques, then to provide approximate answers for many different classes of aggregate queries. In this paper, we present D-skyline and T-skyline methods give almost “true” results on approximated analysis for similarity search query in streaming environments.

This study investigates the potential impact an ethernet bridge based Network Intrusion Detection System (NIDS) would have upon network performance. Three operating systems( OpenBSD, FreeBSD and Linux) were used to test three bridges. The impact of these bridges on network performance while running a NIDS Snort in various modes of data capture was evaluated. The results from this study reveal that an operating system bridge supporting a Network Intrusion Detection Systems sensor will have a significant impact upon network performance.

Streaming media technology has seen tremendous growth and is currently a popular research area. While most of the work on streaming media technology is on ensuring the quality of the video playback and the scalability of the overall media streaming solution, there are few researches that are being conducted to address the limitations of the network, computer hardware nor the streaming media characteristics. These are the other factors that causes network capacity bottleneck. It is generally accepted that a Peer-to-Peer network is suitable for streaming media network and these can be classified into two main network architectures; a tree- based architecture and a mesh-based architecture. In this paper, we investigate and evaluate the network limitations between a Peer-to-Peer tree-based architecture and the more popular mesh-based architecture for media streaming performance. The simulations were conducted under various real-world scenarios and evaluated using different critical performance metrics that affects the reliability of the streaming quality and performance.

Navigation systems have become an interesting research and industrial topic due to their wide deployment. The method of updating the map database (DB) with previously unknown map information has become an important factor in navigation system design. At present, multiple map DB providers maintain their own datasets and work to reduce unknown map information in only their proprietary maps. Map updating systems can generally be divided into two fields: position and traffic sign detection, and network and cloud based map generation and updating. This paper introduces the main difficulties associated with determining and updating unknown map information and briefly surveys the state-of-the-art technologies in this field. Further discussion on potential trends in the development of unknown map information is also provided.

Web services need to be orchestrated to achieve complex demand. However, existing approaches on Web service composition are mainly based on syntax match. It is very difficult to compose heterogeneous Web services with syntax in accord with function. Gradually, Petri net becomes a powerful system modeling and analysis tools research discrete event dynamic system. In order to solve this problem, we present a method to compose Web services automatically based on generalized stochastic Petri nets. Firstly, the generalized stochastic Petri nets (GSPN) and OWL-S composition process are used to model the automatic composition process, and then the Web service composition is realized. Finally, the feasibility of this method is demonstrated though the analyses of a case study with a booking service system. Compared with the existing methods of Web service composition, this method can describe Web service automatic composition more effectively.

Recently, shortest path tree construction is essential in network routing. Dijkstra algorithm, one of the static routing algorithms, is widely used. When some links develop new weights, dynamic routing algorithms become more efficient than static routing algorithms. This is because dynamic routing algorithms reduce the redundancy caused by re-computing the affected part of the network in regards to the changed links. However, dynamic routing algorithms are not always efficient in some cases and increase the computation time when making the shortest path tree. In this paper, we present a Hybrid Shortest Path Tree (HSPT) algorithm which reduces the total execution time of shortest path tree computation by using the advantages of both static and dynamic routing algorithms. Comparisons with the other routing algorithms such as Dijkstra, Dynamic Dijkstra and RDSP show that the HSPT algorithm provides a better performance as demonstrated by the decrease in the execution time.

In digital video broadcasting-handhelds (DVB-H), the contents to be transmitted come in the form of internet protocol (IP) datagram, which are passed to the multi- protocol encapsulation with forward error correction (MPE-FEC) entity. MPE-FEC employs erasure Reed-Solomon (eRS) error correction combined with cyclic redundancy check (CRC), is defined in standard. However, a critical factor must be considered when dealing with a long-term fading channel. If there is only one bit error in an IP packet, the entire IP packet is considered unreliable, even if it contains many correct bytes. To mitigate this effect, this paper proposed hybrid decoding algorithm using the log likelihood ratio (LLR) values. By simulation we show that the performance of the proposed decoding algorithm is superior to that of conventional one.

To reproduce binaural sound at a listener’s ears using stereo loudspeakers in a portable device environment, it is important to design inverse filters that cancel out acoustical crosstalk signals. However, the direct and the crosstalk paths of a head- related transfer function are similar at certain frequencies, resulting in an excessive boost at those frequencies when inverse filters are designed. To mitigate this problem, a cross-talk cancellation filter design method is proposed which allows for the selective attenuation of unwanted peaks in the spectrum by constraining the magnitude of the difference between the direct and crosstalk paths. The performance of the proposed method is then evaluated by means of subjective source localization and objective tests. It is shown from the tests that the proposed method is able to provide binaural sound effects with very closely spaced stereo loudspeakers.

Transportation cyber physical systems such as automotive, aviation, and rail involve interactions between software controllers, communication networks, and physical devices. These systems are among the most complex cyber physical systems being designed by humans, but added time and cost constraints make their development a significant technical challenge. Formal specification technologies are now indispensable for quickly developing safe and reliable transportation systems. In this paper, we propose a formal specification approach for Transportation cyber physical systems. The proposed formal framework is such a formwork. On the one hand, it can deal with continuous-time systems based on sets of ordinary differential equations. On the other hand, it can deal with discrete-event systems, without continuous variables or differential equations. We present a combination of the formal methods Timed-CSP, ZimOO and differential dynamic logic (DL). Each method can describe certain aspects of a transportation cyber physical system: CSP can describe communication, concurrent and real-time requirements; ZimOO expresses complex data operations; differential dynamic logic (DL) model the dynamics and control (DC) parts. A case study of train control system illustrates the specification process for Transportation cyber physical systems.

Beamforming is a signal processing technique that enables antenna arrays to create directional signals, increasing transmitter or receiver gain. We propose a new adaptive user antenna beamforming technique for Multi-cell Cooperative Networks which simultaneously communicates with multiple available BSs and RSs using the MS ’s multiple antennas. We show that the proposed adaptive beamforming technique outperforms distributed beamforming by increasing the data rate with less degradation of the BER.

In this paper an example for the dynamical model for gamification is considered and simulated. Based on the theories of Game Design Features (GDF - cool features, fancy graphics, challenging puzzles, and an intriguing setting and story), Key Characteristics of a Learning Game (KCLG - Challenge, Curiosity, Fantasy and Control), a theory of educational environment design model (ARCS - attention, relevance, confidence and satisfaction), and the theoretical background of Gamification labeled as the MDA (Mechanics, dynamics, and aesthetics) framework, four primary factors such as curiosity, challenge, fantasy and control have been originated. By using these four primary factors, the dynamical model for Gamification was developed and simulated. Furthermore the sensitivities of the four primary factors are considered for different individuals and simulations are represented. The examples and simulations make the values and characteristics of the four primary factors meaningful.

Because of heavy media multitasking and the possibility of youngsters with current smart and handheld mobile computing devices, many controversies exist on the effects of media multitasking on an individual’s learning and behavior. To analyze the effects of media multitasking on youngsters, we have applied eye tracking technology to see their switching patterns between a primary task and interference ones. We also observed their performance given the task of taking an e-learning course while interference media such as web surfing and chatting were available. The analysis results show that frequent task switching occurs in High Media Multitasking (HMM) groups, and their performance was far below that of Low Media Multitasking (LMM) groups in the task of taking an e-learning course. From these results, it can be said that media multitasking definitely interferes with the primary tasks performance, and needs to be avoided for better performance.

Language Lab have been playing as one of the most crucial role in foreign language school. With the crucial role in the school, how to distribute the inexpensive system is more and more considered. As the REST Technology increasing function and security, the dream of creating ubiquitous open contents web-based language laboratory is increasing in its possibility. In this paper, based on the research of the REST-based Web Service, we propose an Open Contents Web-Based to be implemented using Protocol of REST Web Service.

The growing demand for advanced mobile multimedia let the wireless mobile internet access be developed rapidly. The current mobile network is capable of providing flexible quality of service mechanisms including exact estimation of network bandwidth. In this paper, we propose a new GOP level rate control algorithm which can accurately estimate an initial QP according to the estimated network bandwidth. In H.264/AVC, the first frame of a GOP is encoded in intra mode which generates a large number of bits. The amount of bits for I-frame affects the qualities of the following frames of a GOP since they are encoded using the remaining bits of the bits allocated to the GOP. In addition, the first frame is used for the inter mode encoding of the following frames. Thus, the initial QP affects the following frames as well as the first frame. In the proposed algorithm, we observe the relation between the initial QP and the number of bits of I-frame in a GOP. Based on the observation, we propose a new method to estimate the optimal initial QP which maximizes the PSNR of a GOP. It is shown by experimental results that the proposed algorithm predicts the optimal initial QP accurately and thus achieves better PSNR performance than that of the existing algorithm.

In this paper, we present a novel method for improving contents recommendation accuracy using LBS-based users viewing path similarity. We have previously presented a user similarity-based contents recommendation algorithm using NFC. However, the existing research might not recommend contents related with exhibits which users did not tag but do like because it uses only the information that users tagged the exhibits. Also, it can decrease the quality of service (QoS) because it uses tagging patterns of users that were not interested in the exhibits but tagged them without care and thought. In this paper, to solve these problems of our existing service, we divide an exhibition into the areas through analyzing the wifi signal strength and analyze the areas where the user stays long by using LBS and measure the similarity based on the viewing path between users. By using this analyzed information, the proposed service can recommend contents related with exhibits which are the user`s favorite, but not tagged by the user. Also, it might prevent the degradation of QoS of the existing service because it uses the above mentioned information and the measured similarity.

By demand of expression, a double negation sentences are used in many situations. According to characteristics of double negation sentences, a paraphrase method based on templates is used. The method analysis sentence structure of double negation sentences, extract syntax frame and construct paraphrase templates. The syntax structure of double negative sentence and affirmation sentence are suitable for paraphrase. By comparing the syntax structure of double negative sentences with their affirmation sentences, paraphrase templates are abstracted. Matching of paraphrase templates with paraphrase sentences is completed by calculating their similarity. To evaluate performance of the method, small-scale experiments have been done and got the coverage rate of templates and precision of paraphrase 98.9% and 84.7% respectively. The experimental results indicate the availability of the proposed method, and a small corpus resource is established that it can be provided for further research of sentence paraphrase.

In the area of speech signal processing, real background noise is important problem for noise reduction, therefore more skillful methods are required in this area. Accordingly, this paper proposes a spectrum recovery algorithm using a signal-to- noise ratio classification method based on a classification of a voiced or unvoiced signal. Therefore, the proposed algorithm recovers a speech spectrum from a noisy speech spectrum using a time-delay neural network for noise reduction. As such, the proposed system detects the voiced and unvoiced signal, then reduces the noise spectrums for each input frame using the time-delay neural network. Based on measuring correct classification rates and spectrum recovery results, experiments confirm that the proposed algorithm is effective for speech degraded by various noises.

With the dawning ubiquitous computing age, increasing online-based multimedia data presents new challenges for storing and querying large amounts of data to online recommendation systems. Recent studies on recommendation systems show that graph data model is more efficient than relational data model for processing complex data. This paper proposes a new graph data storage model for the collaborative filtering-based recommendation system. Our proposed storage model efficiently filters out vertices which could not impact on calculating top-k recommended items in collaborative filtering algorithm. We present our structure, mechanisms and experimental results for improving the performance of recommender systems. For showing that proposed mechanisms are applicable in multimedia applications, we use real data set of the online site, MovieLense in the experiment. The result of the experiment shows that proposed approach is efficient storage model for recommendation system.

This paper introduces a PWTMCP (Predictive-Wakeup Tree-based Multi-Channel Protocol) which is a new pattern multi-channel protocol in combination with the traditional TMCP and duty cycling power control technology. Firstly, it divides the wireless sensor networks into several sub-trees based on the top of base station and allocates a channel to each sub-tree ,then information flows transmit along the corresponding sub-tree, which eliminates the interference between the tree and avoids the time synchronization problem. Secondly, it uses duty cycling mechanism, which makes the sender accurately predict the receiver’s wake up time, thereby reducing the energy consumption of sensor nodes. This paper also adopts the special prediction error correction mechanism that effectively addresses timing challenges such as unpredictable hardware and operating system delays and clock drift. Using prediction-based retransmission ensures the retransmission efficient. Finally, we compare the PWTMCP with the traditional TMCP in the aspect of energy consumption and efficiency of communication, in conclusion that PWTMCP performs better in the dense networks.

This paper presents buffer replacement schemes for Solid State Drives (SSDs). SSDs consist of multiple NAND flash memory chips, and deploy internal RAM or NVRAM to maintain the mapping table of flash translation layer. The available internal RAM can be used as buffer to absorb a portion of the read/write requests. In order to maximize the effect of the buffer, we have to increase the buffer hit ratio with generating NAND-friendly access pattern. The presented buffer replacement schemes prioritize hot pages that are likely to be re-written in order to increase the buffer hit ratio. In order to generate NAND-friendly write pattern, the presented scheme perform a replacement in a NAND block unit, because NAND-based storage delivers a good performance when the average write request is large in size. The trace-drive simulation shows that the presented schemes deliver a good performance.

Due to the oil crisis and the insecurity of nuclear energy, the exploitation of renewable energy has attracted researchers’ attentions recently. In particular, to charge electric power in smart grid cities, it is important to integrate multiple renewable energy sources efficiently. Unlike the existing energy sources, the renewable energy sources such as wind and solar have low availability because they are uncontrollable. Thus, in order to make stable and predictable electric power generation, many forecasting models have been proposed. This paper examines whether there is any spatio-temporal correlation between wind speed and sunlight, which is not considered by the existing forecasting models for predicting power generation. To achieve this research objective, we categorize and analyze Jeju weather data by regions and seasons. Through the analysis, we show how wind and solar can complement each other. We expect that our analysis scheme can be used as a basis for predicting power generation in smart grid cities where multiple renewable energy sources are combined together.

Feature extraction and tracking is one of the core Augmented Reality (AR) technologies. A markerless AR provides natural synthesis of real-world and virtual objects as it is able to identify objects directly within a video and obtain relevant information. However, disadvantage of the markerless AR is that it may not be appropriate for resource-constrained devices, such as mobile phones, due to the considerable amount of computations necessary. In this paper, we propose a method to address this problem by putting into multimedia content the metadata necessary to provide AR services, such as virtual object information and supplementary information required for on-screen display. In order to show the feasibility of our approach, we extended the ISO base media file format that various well-known media formats such as MP4 are based on.

Video Surveillance as a Service provides service to users using Cloud technology. The DRAGON-Lab supports an opportunity for researchers to use Cloud technology on an open platform, but the services utilize schemes have not been defined. This paper proposed two VSaaS working models, market model and tender/contract model. A comparision testing are given and proven the proposed models are applicable for the DRAGON-Lab, and the market model evaluated have better performance than the tender/contract model. Both of them are justified as reasonable approaches which fulfiled the lacking of this research field.