Earticle

Virtual fitting system is the system that coordinates the clothes that the customers want on their virtual avatars and it can be the representative interactive digital signage, and the vir-tual fitting systems that adopt the augmented reality method are researched currently. That the information on the products of the brands that adopt the system can be checked only, the coordination with other brands or other clothing products cannot be carried out and it has the problem that the personalized coordination information cannot be acquired because the coordination between the existing clothing products and newly-purchased clothing products cannot be performed. in order to coordinate the clothing product using the existing system, basic coordination knowledge or the information like recent fashion trends should be secured, however it is difficult to perform the proper coordination now that most of users do not pos-sess appropriate coordination knowledge in general. Therefore In this paper, virtual reality-based interactive mirror system that provides the personalized coordination information was suggested to solve the problem of existing interactive mirror system that performs virtual coordination work.

Korean governments have established the infrastructure of technology evaluation as one of ways to promote technology transfer and commercialization because of fast growing demands of technology evaluation. Thus, they are in the middle of developing a web-based automatic evaluation tool called K-PEG and operating it in order to secure the speed and the objectivity of technology evaluation. This paper investigates the structure of K-PEG, the representative online automatic evaluation system of patents, and diagnoses its performance qualitatively. The paper also explores ways of K-PEG improvement by presenting qualitative evidences for them on the basis of both in-depth interviews and focus group interview. The concluding remarks focus on management implications to suggest strategies for both overcoming the intrinsic weaknesses of K-PEG and making a step forward.

Software is naturally intangible and abstract which makes the understanding task difficult. There is a growing need for visualizations that improve the comprehensiveness of its structure, behavior and evolution. Graphically visualizing abstract concepts provides a way to raise the abstraction level and therefore, to reduce the software complexity. The graphical visualization has an important contribution by presenting the software under an abstract synthetic view that gives a quick idea of its content, logic, structure and its entities' relationships. It is widely accepted that it can represent a valuable support during the development and maintenance processes. As AspectJ is a relatively new language with powerful specific constructs, it deserves support tools to visualize its software systems. This paper presents our recent work in software visualization with respect to analyze and visualize the AspectJ software structures using graphical elements well-known from daily life such as the Polymetric View and the City Metaphor to conduct various powerful analyses and permit an intuitive understanding of a given visualization and therefore, to get quickly an overview of a huge and complex software. VizzAspectJ-2D and VizzAspectJ-3D are two tools support we have built on top of the Eclipse platform respectively for the 2D and 3D visualizations.

We extend the use of the least squares method to develop a recursive algorithm for the design of adaptive transversal filters such that, given the least-square estimate of a vector of a filter at iteration, we may compute the updated estimate of this vector at iteration upon the arrival of new data. In this paper, we propose a new tap-weight-updated RLS algorithm for an adaptive transversal filter with data-recycling buffer structure. We prove that the convergence speed of the learning curve of an RLS algorithm with a data-recycling buffer is faster than existing RLS algorithms at mean square error versus iteration number. Also, the resulting rate of convergence is typically an order of magnitude faster than the simple LMS algorithm. We show that the number of desired samples should be increased to converge the specified value from the three-dimensional simulation results of mean square error according to the degree of channel amplitude distortion and data-recycle buffer number. This improvement of the convergence performance is achieved at B times the convergence speed of the mean square error increase in the data recycle buffer number with new proposed RLS algorithm.

Framework is a new programming paradigm which makes it possible to reuse the software architecture. Framework defines the wired-in interactions the default behavior of common applications as well as well defined components. Framework even provides a customizing mechanism. In this paper, we have developed an easy-to-customize framework to enhance the reusability. The aim of our customizing environment is to store the meta information in the framework repository to support the reuse of our MAM (Mobile Assets Managements) framework. Our framework repository contains the structural characteristics and the behavior characteristics of components, as well as interactions among the components. In this paper, the framework-customizing toolkit for mobile applications has been constructed. It helps the process of the framework reuse to be systematic by supporting the following: (1) understanding the framework itself, (2) refining the components, (3) changing the related components automatically, and (4) configuring the mobile applications.

Phenomenal growth in air traffic across the world is leading industries to maximize the use of airspace. One of the promising approaches would be to optimize the utilization of information derived from existing data. In early period, all air traffic rules can be programmed and integrated to the system using a general-purpose language like C#, C++, or Java. To increase sustainability of the system, general-purpose languages are too complicated for the users to understand and maintain. In addition, general purpose languages have to be recompiled every time the code has been changed. This research has developed a domain specific language called Aeronautical Rules Script Language (ARSL). The language is particularly designed for collaborative environment aiming at determining separation minima required between aircrafts at planning phase. ARSL can be used as a formal language for configuring air traffic rules and information sharing. As a domain specific language, ARSL is easy to understand and maintain for aeronautical domain experts. The language has been implemented and integrated into the Collaborative Decision Making (CDM) project of Aeronautical Radio of Thailand to help integrate the major elements essential for defining safety longitudinal aircraft separation. The evaluation results show that the implemented ARSL and its services function accurately, providing the same answers as obtained from the specialist.

The existing code generations methods focus on UML Class diagram, which easily represents code structure such as class, method, attribute, but just possibly generate a skeleton code. In this paper we describe to apply UML Message Sequence Diagram (MSD) for representing interactive behavior among objects, and generate more sophisticated Java Code for Android Platform. We also propose code generation method based on Meta Object Facility (MOF) using model transformation technique. And we show metamodel of MSD and model transformation rules written by Acceleo. Using proposed method, we can optimize Java code of Android platform, and increase more code generation rate than the previous approaches.

In this paper we extend the reliability constrained cost minimization (RCCM) model by Helander et al. (1998) from two view points: time non-homogeneous property on software failure-occurrence process and gap between testing and operational phases of software product. The expected cost minimization with reliability constraint is formulated as a non-linear minimization problem under alternative scenario on the operation. We develop an effective optimization algorithm based on the Kuhn-Tucker conditions and provide an illustrative example on how to design a component-based software.

The pay-per-view (PPV) TV industry is growing fast. The purpose of PPV is to earn revenue from subscribers. However, due to illegal accessing issues, the providers come up with protection method. The protection process from piracy is studied in this paper. The proposed method is based on frequency domain. The simulation results confirm that the proposed method is efficient for scrambling contents in frequency domain.