Earticle

Wireless sensor networks have recently emerged as an important computing platform. Energy conservation and maximization of network lifetime are commonly recognized as a key challenge in the design and implementation of wireless sensor networks. Clustering provides an effective method for prolonging the lifetime of a wireless sensor network. In this paper, we propose an Energy Efficient Cluster-Chain based Protocol (ECCP) for wireless sensor networks. The main goal of ECCP is to distribute the energy load among all sensor nodes to minimize the energy consumption and maximize the network lifetime of wireless sensor networks. ECCP organizes sensor nodes into clusters and constructs a chain among the sensor nodes within cluster so that each sensor node receives from a previous neighbor and transmits to a next neighbor. Furthermore, ECCP improves the data transmission mechanism from the cluster heads to the base station via constructing a chain among the cluster heads. Simulation results demonstrate that our proposed protocol significantly outperforms LEACH, CBRP and PEGASIS in terms of network lifetime, stability period, instability period, energy savings, balancing energy consumption among sensor nodes and network throughput.

Due to “Smart Big Bang” in recent, the applications of convergence industry between IT technology depending on smart-phone based mobile communication networks and others are expanding. Various technologies in other fields are adapted to smart-phone efficiently so as to support more convenient daily life service to smart-phone users. However, the developing speed of products is too fast to following most of issues of current smart-phone applications. The designing and developing techniques of smart-phone system become more complicate for adapting new technologies of smart-phone into wireless communication network. There is no perfect solution to solve these problems. Thus, we propose an Augmented Reality (AR) based embedded integration prototyping for the networked system management as one of method. In particular, we describe a network based ticket reservation system and its organization based on smart-phone applications with augmented reality (AR). Prior to developing and designing real smart-phone application, this embedded integration prototyping is helpful to control and predict operational process in big events such as Film Festival, Biennale, EXPO, and so on. Finally, we apply our proposed method into a certain specific zone such as YEOSU EXPO in order to build a location based real time reservation system.

Augmented Reality (AR) is a view of a real-world mixed with and/or superimposed by virtual objects. Therefore, one of the most important key techniques of implementing an AR is recognizing the environment so that it can correctly determine the virtual objects to be displayed on the multimedia representing the real-world. This paper reviews the “Method of Recognizing Objects in a Picture for Mobile AR Using Electronic Maps.” Then we describe the implementation of the method and an application on an Android mobile in detail.

Password-based authentication scheme is one of the efficient authentication mechanics to protect resources from unauthorized access. Chang-Lee, in 2008, proposed a password-based mutual authentication scheme to overcome the security drawbacks of Wu-Chieu’s scheme. In this paper, we have shown that Chang-Lee’s scheme is vulnerable to various attacks known by literatures. Also we proposed an improved scheme to overcome the security drawbacks of Chang-Lee’s scheme. As a result of analysis, the proposed scheme not only withstands the various attacks, such as the user impersonation attack, the server masquerading, the man-in-the-middle attack, the off-line password guessing, the insider attack, but also provides mutual authentication between the user and the server. At the same time, the proposed scheme is more efficient than the related schemes in terms of the computational complexities.

Many soccer fans prefer to watch a summary of football games as watching a whole soccer match needs a lot of time. Traditionally, soccer videos were analyzed manually, however this costs valuable time. Therefore, it is necessary to have a tool for doing the video anal- ysis and summarization job automatically. Automatic soccer video summarization is about extracting important events from soccer matches in order to produce general summaries for the most important moments in which soccer viewers may be interested. This paper presents a machine learning (ML) based event detection and summarization system for em- phasizing important events during soccer matches. The proposed system rstly segments the whole video stream into small video shots, then it classies the resulted shots into dierent shot-type classes. Afterwards, the system applies two machine learning algorithms, namely; support vector machine (SVM) and articial neural network (ANN), for emphasizing impor- tant segments with logo appearance with addition to detecting the caption region providing information about the score of the game. Subsequently, the system detects vertical goal posts and goal net. Finally, the most important events during the match are highlighted in the resulted soccer video summary. Experiments on real soccer videos demonstrate encourag- ing results. The proposed approach greatly reduces workload and enhances the accuracy of summarizing soccer video matches with reference to both recall and precision performance measurement criteria.

Most implementations of pairing-based cryptography are using pairing-friendly curves with an embedding degree k ≤ 12. They have security levels of up to 128 bits. In this paper, we consider a family of pairing-friendly curves with embedding degree k = 24, which have an enhanced security level of 192 bits. We also describe an efficient implementation of Tate and Ate pairings using field arithmetic in Fq24; this includes a careful selection of the parameters with small hamming weight and a novel approach to final exponentiation, which reduces the number of computations required. When comparing with the latest implementation available in the research community, ours is 15% faster due to both our selection of efficient elliptic curve parameters and faster multiplication on Fq24. Therefore, it can significantly contribute to most contemporary identity-based or attributed- based encryption or signature schemes whose basic and essential operations are based on paring, known as one of the most time-consuming operations.

The high speed network becomes an important indicator of national power in information society. The bandwidth of LAN and WAN are over the Gbps but bandwidth of subscriber access network is dozens of Mbps. Task Force team for IEEE 802.3ah has accomplished the standardization of EPON which is one of next generation subscriber access networks but it doesn’t still have the bandwidth wide enough to support the new service which demands high bandwidth. Therefore IEEE 802.3av Task Force studied 10G EPON technology. It expanded the updown bandwidth range of 1G EPON 10 times in order to support a next generation multimedia service demanding high bandwidth and which is proceeding the standardization of physical layer in IEEE 802.3av Task Force. This paper has designed the model which can accommodate IEEE 802.1 AVB traffics smoothly in 10G EPON and suggesting the Intra-ONU scheduling model which makes this model operate effectively.

IPTV stands for Internet Protocol Television, and any user with an IP device such as a smart phone can get IPTV service anywhere and anytime as long as the user can access the Internet. Because of this advantage, IPTV is widely believed to be the next killer internet application, and many organizations provide IPTV services. As IPTV becomes popular, formal proofs of IPTV service quality have been attempted. One of them was the minimum cycle time analysis. The minimum cycle time analysis takes a long time when the given system is large. This paper proposes a divide-and-conquer method of minimum cycle time analysis that is much faster than the existing ones.

Cloud is one of the most important atmospheric effects in computer graphics applications, and there have been lots of research results on it. Although they have mainly focused on the modeling and rendering of more realistic clouds, nowadays flight simulation games or even full-scale flight simulators often need a quantity of visually plausible clouds rather than realistic ones. Furthermore, we also need to save graphics designer’s efforts, especially for small production companies. Contrary to the previous works, we aimed at the mass production of clouds, and represent a new method for modeling various kinds of visually plausible clouds with as little effort as possible. These clouds are displayed in real time with low computing power consumption. Based on the hierarchical particles, our system starts from locating relatively large spherical particles in the space. Using these seed particles, our system automatically generates descendant particles to represent details of the clouds. In this way, the visually plausible clouds can be generated with much less effort, while the designer may build up specifically shaped clouds through controlling the seed particle locations. The particle hierarchy also enables us to naturally implement level-of -detail effects on the cloud rendering. In the final rendering stage, our system renders each particle with alpha-blended billboards to achieve fast real-time processing.

In this paper, we propose a task-parallel programming extension for the JPEG decoder of the libjpeg-turbo library. Efficient JPEG decoding is especially important for resource-constrained mobile devices such as smartphones, where decoding (e.g., browsing of web pages containing images, image search aso) is far more common than image encoding. The aim of our work is to utilize multiple CPU cores for JPEG decompression from a single client thread. Our method is orthogonal to libjpeg-turbo's support for data-parallelism (SIMD). Experimental evaluation of our approach on a 4-core Intel i7-2600K CPU shows speed-ups of up to 2.5x over the sequential, and up to 34% over the SIMD-version of the libjpeg-turbo JPEG decoder.

Moving objects in indoor space has been a research focus in recent years, as most people live and work in indoor space, e.g. working in office, living in apartment, etc. In this paper, we make a first step in indoor moving object management. We focus on the conceptual modeling of indoor space as well as indoor moving objects, and aim to describe the semantics and properties of indoor moving objects. Firstly, a conceptual modeling framework for indoor space is defined, based on which we propose a semantic description of indoor moving objects. Compared with previous models, our model takes into account the relationships among rooms, doors, sensors and moving objects, and uses a layered approach to represent indoor space and indoor moving objects. The model proposed can be further extended to meet different needs in indoor moving object monitoring and tracking.

This paper introduces a hybrid approach that is based on color information that utilizes a mask and K-Means clustering algorithm along with the frame averaging background subtraction technique. This hybrid approach efficiently removes artifacts caused by lightening changes such as highlight and reflection from segmentation, while also successfully removing shadows of stationary objects and dark cast shadows. Dark cast shadows cause an issue with tracking and detection. To eradicate these shadows, we first create a mask by assigning values to R, G and B channels utilizing the shadow properties to this RGB color individually, and then we apply K-Means clustering algorithm to this mask for efficient removal. Simulation results from several video sequences with different scene conditions also reveal the effectiveness and robustness of the proposed algorithm.

In this paper, a situation-based storyline generation system is proposed. The generated new plot is a modification of situations stored in situation database, which is different from previous character-based and story-based systems. The new plot is generated based on the current plot with rules applied from the constraint database. The number of possible plots will be huge by applying this method. After the generation stage, a director module will check if the plot is the end of the story. We discuss detailed process modules as well as the situation and constraint database.

A new side information update method of error correction decoder for distributed video coding is proposed. This method improves the quality of side information. The side information values are updated using both the values of quantization index and its quantization interval. Side information update method reduces the amount of parities transmitted to decode quantization index.

Due to the dynamic nodes’ movement and their interaction, it is challenged to represent, discover and exchange location information in a ubiquitous environment. In this paper, we present a new approach to location information awareness that organizes the format of nodes’ location information and maintains the communication between direct nodes in a ubiquitous environment. The structure of Context information Base (CiB) and Context information Communication protocol (CiComm) is manipulated to implement the proposed features.

This paper proposes LED driver to minimize power consumption due to LED forward voltage difference and temperature rising. Compared to conventional LED driver, the proposed driver had excellent stability, brief structure and linear output voltage of DC-DC boost converter. With the channel current set at 100mA, the on/off switchover was controlled by means of the PWM control signal in the simulation and actual installation. As a result, it turned out that the normal LED current value of 100mA was produced according to the input PWM control signal.

Range-based Probabilistic Group Nearest Neighbor (in short RP-GNN) query has recently gain much attention, due to its wide usage in many Location Based Services (LBSs). Previous works mainly focus on the uncertainty of data objects (P). While the uncertainty of query objects (Q) is prevailing in reality. In this paper, a comprehensive discussion on uncertain query objects is presented. Meanwhile two novel pruning methods are proposed to improve the performance of RP-GNN: one is Query points pruning (Q_pruning) and the other is Geometric pruning (G_pruning). Q_pruning reduces the number of query objects needed to be considered. And G_pruning method exploits the geometric properties of the RP-GNN problem to narrow down the search space. Extensive experiments show the effectiveness, efficiency and scalability of proposed methods.

A single sign-on scheme is proposed in which a user offers his credential information to a home network running the OSGi (Open Service Gateway Initiative) service platform, to obtain user authentication and control a remote device through a mobile device using this authentication scheme, based on SAML (Security Assertion Markup Language). Single sign-on profile is defined to overcome the handicap of the low computing and memory capability of the mobile device and automated user authentication is applied to control a remote media device using a mobile device in a ubiquitous home network based on the OSGi.

Transportation volume of global container is increasing as trade grows larger than before based on lifting trade barriers. At the same time, participants using logistics service require real-time based information of container location, security for preventing a robbery case, and temperature/humidity/impact. But existing system based on bar code, EDI, RFID technology is possible to communicate only part of point in all logistics network. In addition, constructing infrastructure cost high because logistics network is based on all over the world these days. Container tracking system that is suggested in this research is one of the methods for realizing Global SCM(Supply Chain Management). This would provide feature for managing container as communicating data between tracking device and information gathering server based on real-time. This research provide communicating method which has systematic communication standard and data format for diverse logistics companies when real-time based tracking device is communicating with location information gathering server.

We present a dynamics model of rotor blades for real-time helicopter simulation. Collisions between the air flow and the moving blades make helicopters fly. In aerodynamics, or even in computer simulations, they precisely analyzed the collisions between the fluid(air) and the solid object(blades), and calculated the differential equations from the collisions. Thus, it was hard for them to generate real-time helicopter motions due to massive computations for calculating the equations. In this paper, we start from a geometric model of rotor blades, which reflects the characteristics of real world blades due to the various factors from helicopter aerodynamics, although some factors should be simplified to show real-time behaviors. Based on this geometric model, we present a dynamics model for calculating the forces due to the rotor blades colliding with air flows. Our dynamics model interprets the collisions between the fluid and the solid objects as the action-reaction forces, as originally Newton did. Finally, we present the force equations suitable for the existing rigid-body simulation systems, instead of fluid-dynamics equations. We implement a prototype system for helicopter motions, and it shows sufficient real-time processing behavior with ordinary PC’s.

A location-based service (LBS) cannot be realized unless solutions of the positioning problem are available at hand. For the outdoor positioning, GPS based practical solutions have been introduced. Using GPS they have developed so many commercial LBS systems. Navigation, logistics, troop management and fleet management are all examples of LBS. LBS is so useful that it should be available in doors. However, GPS signal is so weak inside buildings that we cannot determine the location of a moving object in doors with GPS only. Therefore, so many indoor positioning researches have been performed. Cricket, Active Badge and BAT are pioneers in the field of indoor positioning. They are very accurate but they require special equipments dedicated for positioning. Using special equipments is not economical. Therefore, many researchers have suggested using wireless local area networks (WLAN) in positioning. Among the methods they are using, the fingerprinting methods are most accurate. The deployment of the fingerprinting methods consists of two phases: the off-line phase and the on-line phase. During the off-line phase a site-survey of the received signal strength indices (RSSIs) from access points (APs) is performed. The vector of the RSSI values at a point is called the location fingerprint of that point. A lot of location fingerprints must be collected at each of the points in the site during the off-line phase. This is extremely tedious and time consuming. An alternative choice is the trilateration method. This method converts RSSIs from APs into distances and determines the location of the moving object with the distances and the locations of the APs. That is, we only need the coordinates of the APs in the site to get ready to run the trilateration positioning program. The conversion rule of RSSIs into distances is based on the RF propagation loss model. The model is a simple mathematical expression representing the relationship between the RSSI and the distance. However, the RSSI is influenced by obstructions, reflections and multipath and the RF propagation loss model is very erroneous. As a result, WLAN based trilateration is much less accurate than the fingerprinting method. Nevertheless, the trilateration method could be more practical than the fingerprinting method because it does not require the time consuming off-line phase process. Therefore, they established IEEE 802.15.4 A where the distance is determined by the speed of RF and TOF (Time of Flight). Ubi-nanoLOC mote complies with IEEE 802.15.4 A. This paper discusses the advantages and disadvantages of the WLAN-based trilateration indoor positioning and the Ubi-nanoLOC indoor positioning draws our final conclusions.

Recently DRAM and PRAM hybrid main memory organization has been studied in order to address the high levels of energy dissipation in DRAM based main memory. It is expected that this new memory architecture will be used soon in mobile computers which use NAND Flash memory based storages. In such computers, legacy operating system functionalities like file system and memory system should be modified in order efficiently to manage heterogeneous memory organization. In this paper, we study a new buffer cache scheme which considers DRAM/PRAM hybrid main memory and flash memory based storages. The goal of proposed buffer cache scheme is to minimize the number of write operations on PRAM and the number of erase operations on flash memory while maintaining the cache hit ratio. In order to evaluate proposed scheme, we performed trace-driven simulation.

Embedded systems are suitable for application system with strict requirements on the functionality, reliability, cost, size and power consumption Also embedded Linux has been widely used to its numerous characteristics, such as open-source, small size, fast speed, and excellent network performance. Therefore, embedded Linux gets more attention and has more application in industrial monitoring field. In this paper, we propose system development process based on embedded Linux. Using this system development process can increase the reusability of development process and time.

This study was designed the physical signal control system of posture balance for preserving body sway. It was necessary to constantly monitor their physical parameters, especially body balance, to effectively control their health condition and care body moving. There signified body movement by the parameter that was neck moving and eye moving such as forward/backward body moving. This system consist of a data acquisition system, which was responsible for capturing and sending data signal, which was responsible for gathering data signal processing, and to control the received data through feedback system. There were evaluated their body sway through a physical data, and identified to maintain the stability. It was expected to control a body activity and to confirm the physical signal system.

In this paper, we present a novel object detection scheme using the multi-scale contour fragments. The template fragments are extracted by decomposing the template contour. The multi-scale hinge angle, contour direction and partial Hausdorff distance (PHD) are used to select candidates in the edge image. Then, the matches with different scales and directions are selected by the Multiclass Discriminative Field (MDF) from the candidates. With the matches and their corresponding sample fragments, the contours of the objects can be obtained. The experiments on our postmark dataset and the ETHZ dataset show that the proposed scheme is robust to detect a class of objects with different scales, directions and complex background.

The blooming of location-sharing services has led to serious privacy concerns, particularly location privacy concerns. In this paper, we first discuss the definition of location privacy and then review several location privacy protection mechanisms. After that, we classify location privacy protection mechanisms into four categories: legislation & regulatory, anonymity & obfuscation, protocol and policy-based protection. We explain each privacy protection mechanism category in detail and discuss if it can be used in particular kind of location-sharing services along with their advantages and disadvantages.

The channel allocation is one of the key problems in the design of the wireless network, as it greatly influences the throughput and performance of the network. Firstly we introduce the channel propagation model, which is able to accurately reflect the signal attenuation, the network interference and other factors in the actual environment. Based on this work, we put forward the channel allocation algorithm named LRCAA, which includes two parts: the pre-allocation stage and the dynamic optimization stage. The LRCAA is verified by the experiment, which shows it can converge to a stable optimal point and obtain a better network performance.

In this paper, we design a low-cost checkpointing-based rollback recovery algorithm to address the traditional scalability problem of synchronous checkpointing in the completely different point of view compared with existing ones. This algorithm enables a cluster-wide set of processes to take their semi-global checkpointing procedure while a small set of cluster heads monitor local commit of their respective administrative areas and always observe the global consistency condition. It can considerably lower communication overhead that may occur in the previous ones. This feature can enormously decrease the frequency of cluster-to-cluster communications especially in large-scale hierarchical multi-cluster systems.

NAND-based storage devices deploy the flash translation layer (FTL) in order to emulate the block device characteristics because NAND flash memory does not support the overwrite operation. The FTL schemes that use log blocks such as the BAST and the FAST scheme are adequate for the devices with harsh memory. This paper presents the log block replacement scheme to improve the performance of the FAST FTL scheme. The presented scheme considers the number of valid pages of the candidate log block when selecting the victim log block, because the cost of the garbage collection decreases as the number of valid pages in the victim log block is less. The presented scheme gives the second chance to the candidate log block if its valid pages are more than the threshold. The simulation shows that the presented scheme improves the performance of the FAST scheme up to 5.0 %. The improvement is more conspicuous as more NAND blocks are used as log blocks.

The demand of high volume data communication over the internet is mounting day by day. In general, the performance of a communication system depends on the loss of data packets. It happens as there are multiple paths exist in wireless networks. For the reliable and secure data communication over Internet Transmission Control Protocol (TCP) is playing a significant role. On the development of TCP, a number of approaches are already designed and tested in the communication system. Generally, TCP congestion parameters are set in sender site in a communication system. In this paper, we investigate the download system performance tuning the TCP congestion parameters at the Ethernet port of receiver side. Based on the experimental study, it is concluded that tuning TCP parameters at receiving side improves the download system performance by reducing packet loss, increasing download speed and maintain stable I/O and time/sequence graphs.