Distributed Parallel Processing Laboratory

Nikolay Mirenkov Professor	Rentaro Yoshioka Assistant Professor	Peter Simon Sapaty Visiting Researcher

Existing systems of symbols and notations are usually very abstract and there is a great gap between the form and meaning of data/knowledge. Our long-term education, in an essential part, is reduced to training our brain for being "encoding-decoding" machine bridging this gap. The abstractness mentioned and bridging operations are sources of serious mental and physical problems for a great variety of people and, especially for disabled and elderly. Our aging society is also becoming information society. So, the above-mentioned aspect of our environment is becoming crucial. That is why, our research is to develop a new environment with lesser level of abstraction and to investigate how it can improve the quality of people life. Our general program is cyber-infrastructure including high-performance computing. We are also thinking about active knowledge being developed by humanity and undertaking research efforts in visual (multimedia) languages and tools, parallel and distributed systems. In a great part, our research and development are based on an idea of self-explanatory components in a cyberFilm format. A cyberFilm is a set of color stills supported, if necessary, by text, voice/sound and special links. Each still is to represent a view (some features) of objects or processes. Each cyberFilm is to represent a multiple view (an extended set of dynamic and/or static features) of objects or processes. Different views can be related to different moments of time, positions in space, levels of hierarchy, values of data attached to space points, etc. Different media can be used for different views. A self-explanatory cyberFilm means that the associated stills are organized and presented in such a way that the semantic richness is clearly brought out. The investment of meaning in the cyberFilm is reduced to developing a series of viewswatching (and hearing) in non-linear order. Usually, a still is self-evident and a cyberFilm is a result of special gathering of clues or hints. This result is a piece of knowledge. So, self-explanatory adequacy depends on this knowledge. The more accurate and relevant views are used, the greater adequacy is reached. The idea of cyberFilms is used for the specification of information resources and programming operations with the resources, as well as for the representation of multimedia messages and implementation of human-computer interfaces. The idea of equal opportunities to all individuals in the use of information resources is used to create a right set of cyberFilms and methods of their adaptation. We lead three clusters of projects related to filmification of methods and data: 1) Active Knowledge Studio for teachers, students, and programmers, 2) F-Communication Systemfor children, elderly and handicapped people, and 3) Virtual objects, haptic interface and tactile visualization for medical doctors and caretakers.

Refereed Journal Papers

[nikmir-01:2004]	Tetsuya Hirotomi and Nikolay Mirenkov. Augmentative and Alternative Communication Based on Multimedia Hieroglyphs. Journal of the Japanese Society for Wellbeing Science and Assistive Technology, 3(2):40{50, 2004.
	We have developed an augmentative and alternative communication system oriented to people with expressive disorder. In this system, multimedia is used to fully utilize available communication capabilities of different users. The system is based on film formats ofmultimedia wordsand sentences called multimedia hieroglyphs. Corresponding films are multiple views of objects and/or processes. They are pieces of knowledge. These pieces are acquired in a network-accessible film database. To access them, the system provides special interfaces implemented as rather independent subsystems oriented to different groups of users. These interfaces support keyboard-less operations for searching, creating, and browsing multimedia words/hieroglyphs, as well as sending and receiving corresponding messages. The goal of this paper is to represent a new computer-mediated communication environment. This environment is based on our approach. It includes ideas of multimedia sentences consisting of multimedia words as well as self-explanatory features behind such words. Special attention is paid to the implementation of F-Mail subsystem oriented to cerebral palsied people. Initial results of usability tests are also provided.
[nikmir-02:2004]	Mahmoud Saber and Nikolay Mirenkov. A visual representation of cellular automata-like systems. Journal of Visual Languages and Computing, 15:409-438, 2004.
	In this paper, we describe a new special-purpose programming language developed for visual speci??cation, presentation, and explanation of cellular automata (CA) systems within a visual programming environment, as well as, for programming them. This language is based on using visual patterns, colors, and animation for representing the CA system structures and operations on these structures, and for performing editing and composing manipulations with corresponding software components. Examples of the CA algorithm representations and some details of the environment implementation are presented.
[nikmir-03:2004]	Robert Roxas and Nikolay Mirenkov. A Visual Environment for Specifying Global Reduction Operations. International Journal of High Performance Computing and Networking, 1(2), 2004.
	This paper presents an approach to specifying the different types of global reduction operations without laboriously coding the source code using the traditional textual approach. We use the visual environment provided by the system called active knowledge studio (AKS) to specify, modify, view, or run the specification. As a background, we give an overview of the AKS system and the main technique being employed that is using CyberFilms in building programs. Although the main focus of this paper is to show how to override at the lowest level the default parameters supplied by the system, we also include a sufficient background about the higher levels so that the reader can relate the lowest level with the other higher levels in the hierarchy of specification. In general, this paper discusses how to specify global reduction operations using a language of micro-icons and shows how the visual programming environment supports the manipulation with these micro-icons.
[nikmir-04:2004]	Tsukasa Ebihara, Nikolay Mirenkov, Masaaki Nemoto, and Rikio Nomoto. Filmification of methods and an example of its applications. International Journal of Software Engineering and Knowledge Engineering, 15(1):87-115, 2005.
	A user-oriented programming paradigm based on the 'filmification' of methods is developed by the Active Knowledge Studio team at the University of Aizu. Within this paradigm, an algorithmic 'cyberFilm' is a set of multimedia frames where different groups of frames represent different algorithmic features from multiple views. Because of the multiple views and multimedia, users can understand the algorithm correctly and easily. In this paper, an exampleof the filmification ofmethods is presented, related to cyberFilm frames of the thermal conduction problem solved with the Gauss-Jordan method. Through these frames, features of the algorithm are presented, and an essential compactness of an integrated view, considered as a visual program, is explained. Examples of of program templates supporting the cyberFilm frames are also presented. Two types of program synthesis schemes supporting program synthesis are also described.
[nikmir-05:2004]	Oleg Monakhov and Nikolay Mirenkov. Developing parallel programs in multimedia film environment. Journal of Three Dimensional Images, 18(4):91-95, 2004.
	An environment for manipulating and translating multimedia program skeletons ('films') into parallel programsis considered. Amaingoal of this environment is to make easier the creation of parallel programs for various parallel computing systems.

Referred Proceeding Papers

[nikmir-06:2004]	Yutaka Watanobe, Rentaro Yoshioka, and Nikolay Mirenkov. Library architecture for searching software components by their algorithmic features. In Proceedings of 2004 International Conference of Distributed Multimedia Systems, pages 190-195, San Francisco, USA, Sep. 2004.
	A library architecture suitable for searching software components by their algorithmic features is presented. One of the main goals of the library is to provide an eAEcient searchmechanism for a large number of reusable software components. Toachieve a higher degree of reusability in comparisonwith conventional libraries, a library in which eachcomponent is stored as acyberFilm is considered. A cyberFilm is acomponent format that can represent software as a set of algorithmic features, such as structures, schemes, formulas, and I/O operations. These algorithmic features provide a common, high-level and yet precise description of computation. The library takes advantage of these features in the searching operations. A set of softwaremodules responsible for manipulations with corresponding features constitute the library architecture. In this paper, an overview of the attributes derived from different features to classify and access cyberFilms is presented. Applicability of the attributes to support searching interfaces is also described.
[nikmir-07:2004]	Yutaka Watanobe, Rentaro Yoshioka, and Nikolay N. Mirenkov. A Searching Method Based on Problem Description and Algorithmic Features. In Proceedings of 2005 International conference of Databases in Networked Information Systems, LNCS3433, pages 138-149, Japan, March 2005.
	A searching method related to 'what' and 'how' problem descriptions within a special software component library is presented. The 'what' problem description is based on a high level representation of general features of initial and final data the problem can operate and produce. The 'how' problem description is based on another high level representation of computational features of algorithm used for the problem solution. To realize the searching method, a special library in which each item is stored as a cyberFilm is considered. CyberFilm formats are used for representing data/knowledge units as self-explanatory components, and a special double ID is assigned to each component. The double ID is a pair of a URL-like address and a classification code. A cyberFilm represents a software component as a set of algorithmic features that provide a high-level and yet precise description of computation as well as semantics of problem that should be solved by the corresponding algorithm. These features are the basis of the classification code and the searching method. In this paper, a basic idea of the 'what' problem description for the searching method, and an overview of the features for the 'how' problem description, are considered. A possible library structure that uses the classification codes to support various searching goals is also described.
[nikmir-08:2004]	T. Tsuchimoto, Y. Nishidate, N. Mirenkov, and G. Nikishkov. Augmented Reality Interface for Measurement of Tissue Stiffness. In Proceedings of the Fifth Int. Conf. on Human and Computer, pages 148-153, Aizu-Wakamatsu, Japan, Sep. 2004.
	In this paper, we present an implementation of augmented reality interface for measurements of tissue stiffness using the tactile sensor. A developed augmented reality system provides real-time visualization of tissue stiffness measurements. Head-mounted display, small video camera and notebook computer are employed. Tracking of real-world objects is performed with a marker attached to a patient body. Results of stiffness measurements are depicted as semi-transparent three-dimensional objects projected on the patient body.
[nikmir-09:2004]	P. Sapaty, M. Sugisaka, N. Mirenkov, M. Osano, and R. Finkelstein. Grasping the Distributed Entirety. In Proceedings of The 10th International Symposium on Arificial Life and Robotics, Beppu, Oita, Japan, Feb. 2005.
	In this paper, as an alternative to existing manned or unmanned control systems, we are describing flexible WAVE-WP (or World Processing) ideology and technology aimed at creating new or penetrating other systems and their organizations. Supported by a high-level spatial programming language, WAVE-WP allows us grasp large systems as a whole, study their behavior and direct evolution in the way required. On the implementation layer, the technology widely uses self-spreading mobile cooperative program code dynamically covering and matching distributed systems. This often allows us to get solutions orders of magnitude more compact than by other approaches. The internal system organization, including partitioning intocomponents and specific commandand control, can be a function of the environment and mission scenario; it may change at runtime while preserving the overall system operability and goal orientation.
[nikmir-10:2004]	P.-A. Fayolle, A. Pasko, E. Kartasheva, and N. Mirenkov. Shape recovery using functionally represented constructive models. In International Conference on Shape Modeling and Applications, IEEE Computer Society, pages 375-378, Genova, Italy, May 2004.
	We propose a method which helps to fit existing parameterized function representation (FRep) models to a given dataset of 3D surface points. Best fitted parameters of the model are obtained by using a hybrid algorithm combining simulated annealing and Levenberg-Marquardt methods. The efficiency of the approach is shown for recovery of two test items. We show through the CADmodel processing an application of the proposed approach to the shape recovery followed by finite element mesh generation and adaptation.
[rentaro-01:2004]	R. Yoshioka, Y. Tsuchida, and N. Mirenkov. Visual Editor for Designing and Editing Algorithmic Skeletons. In Proceedings of the Tenth International Conference on Distributed Multimedia Systems (DMS'2004), pages 63-68. Knowledge System Institute, Sept. 2004.
	A new type of multimedia editor for designing algorithmic skeletons in a cyberFilm is presented. A cyberFilm is a new multimedia format to represent and edit computation through a set of algorithmic features. The algorithmic skeleton is one of these features, in which the algorithmic structure and the associated flow of activities are described. The algorithmic skeleton is represented by a series of stills and scenes. Each still consists of multimedia symbols such as images, text and sounds, and corresponds to one computational step. A scene is a subset of stills which can be considered as related steps and meaningful unit of an algorithm. The use of multimedia symbols in the stills and the composition of stills into scenes must follow the syntax of the cyberFilm language. The design of the skeleton is performed by editing these multimedia symbols in the stills and scenes. This process can be extremely simplified by a special editor supporting high-level operations and intelligent interfaces. This editor allows even a novice user to draw a picture conforming to the language and simplifies the task of extracting and transforming the user's computational intention into the algorithmic skeleton images and animations. In this paper, the different algorithmic features constituting the algorithmic skeletons are described and a number of sub-editors to edit those features are presented.
[rentaro-02:2004]	Y. Watanobe, R. Yoshioka, and N. Mirenkov. Library Architecture for Searching Software Components by their Algorithmic Features. In Proceedings of the Tenth International Conference on Distributed Multimedia Systems (DMS'2004), pages 190-195. Knowledge System Institute, Sept. 2004.
	A library architecture suitable for searching software components by their algorithmic features is presented. One of the main goals of the library is to provide an eAEcient search mechanism for a large number of reusable sofware components. To achieve a higher degree of reusability in comparisonwith conventional libraries, a library in which eachcomponent is stored as acyberFilm is considered. AcyberFilm is acomponent format that can represent software as a set of algorithmic features, such as structures, schemes, formulas, and I/O operations. These algorithmic features provide a common, high-level and yet precise description of computation. The library takes advantage of these features in the searching operations. A set of softwaremodules responsible for manipulations with corresponding features constitute the library architecture. In this paper, an overview of the attributes derived from different features to classify and access cyberFilm is presented. Applicability of the attributes to support searching interfaces is also described.
[rentaro-03:2004]	Y. Watanobe, R. Yoshioka, and N. Mirenkov. A Searching Method Based on Problem Description and Algorithmic Features. In Lecture Notes in Computer Science: Databases in Networked Information Systems: 4th International Workshop (DNIS 2005), volume 3433/2005, pages 138-149. Springer-Verlag GmbH, March 2005.
	A searching method related to what and how problem descriptions within a special software component library is presented. The what problem description is based on a high level representation of general features of initial and final data the problem can operate and produce. The how problem description is based on another high level representation of computational features of algorithm used for the problem solution. To realize the searching method, a special library in which each item is stored as a cyberFilm is considered. CyberFilm formats are used for representing data/knowledge units as self-explanatory components, and a special double ID is assigned to each component. The double ID is a pair of a URL-like address and a classification code. A cyberFilm represents a software component as a set of algorithmic features that provide a high-level and yet precise description of computation as well as semantics of problem that should be solved by the corresponding algorithm. These features are the basis of the classification code and the searching method. In this paper, a basic idea of the what problem description for the searching method, and an overview of the features for the how problem description, are considered. A possible library structure that uses the classification codes to support various searching goals is also described.

Grants

[nikmir-11:2004]	N. Mirenkov. Ministry of Education Scientific Research Fund, 2001-2004.
[nikmir-12:2004]	N. Mirenkov. Knowledge Cluster Creation Support Project, 2003-2005.
[nikmir-13:2004]	N. Mirenkov. Fukushima Prefectural Foundation, 2004.
[rentaro-04:2004]	R. Yoshioka. Grant-in-Aid for Scientific Research, Japanese Ministry of Education, Culture, Sports, Science and Technology, 2003-2005.
[rentaro-05:2004]	R. Yoshioka. Subsidy from the Fukushima Foundation for the Advancement of Science and Education, 2004.

Academic Activities

[nikmir-14:2004]	Nikolay Mirenkov, Sep. 2004. Co-chair of the Program Committee, International conference DMS-2004, USA
[nikmir-15:2004]	Nikolay Mirenkov, 2004. Member, the IFIP Working Group 10.3 (Concurrent Systems)
[nikmir-16:2004]	Nikolay Mirenkov, 2004. Associate Editor, the Tamkang Journal of Science and Engineering, International Journal
[nikmir-17:2004]	Nikolay Mirenkov, 2004. Member of ACM and IEEE
[nikmir-18:2004]	Nikolay Mirenkov, 2004. A referee report for Journal of Visual Languages and Computing, Elsevier
[rentaro-06:2004]	R. Yoshioka, Sept. 2004. Programming committee member and session chair, International Conference on Distributed Multimedia Systems

Patents

[nikmir-19:2004]	N. Mirenkov, R. Yoshioka, and Y. Watanobe. Self-explanatory programs and methods of manipulations with them, filed, 2005-56804, March 2005.
[rentaro-07:2004]	R. Yoshioka, N. Mirenkov, and Y. Watanobe. Self-explanatory programs and methods of manipulations with them, March 2005.

Ph.D. and Other Theses

[nikmir-20:2004]	Tsukasa Ebihara. PhD Thesis: A Program Generator from CyberFilm Specifications, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-21:2004]	Robert Roxas. PhD Thesis: Visual Specification of Intercompont Communication, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-22:2004]	Mahmoud Saber. PhD Thesis: Visual Representation and Specification of Cellular Automata-like Algorithms, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-23:2004]	Masaaki Nemoto. Master Thesis: Code Generator fromCyberfilm Specifications and its Error Checker, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-24:2004]	Naohiro Nakamura. Master Thesis: Robots and toys in F-Literacy System, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-25:2004]	Koshi Makino. Graduation Thesis: Convolution Neural Network-Based Face Detection, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-26:2004]	Satoshi Iwano. Graduation Thesis: English Translator and Word Attributes for Multimedia Hieroglyph Tanslation, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-27:2004]	Yoshiko Ogasawara. Graduation Thesis: Random Data Interpolation to Create Datamaps for Tactile Data Visualization, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-28:2004]	Kei Takanashi. Graduation Thesis: A Viewer and Editor for an Integrated View of Algorithmic Features, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[nikmir-29:2004]	Tetsuro Hoshi. Graduation Thesis: Multiple Device Interface for F-Mail System, University of Aizu, 2004. Thesis Advisor: Nikolay Mirenkov
[rentaro-08:2004]	Kei Takanashi. Graduation Thesis: A Viewer and Editor for an Integrated View of Algorithmic Features, University of Aizu, 2004. Thesis Advisor: Mirenkov, N.

2006 THE UNIVERSITY OF AIZU ALL RIGHTS RESERVED