APAN 2003 Conference in Fukuoka
B1: Access Grid Tutorial (Room B 09:00-10:30, Jan 22)
The Access Grid has become for many a standard part of scientific collaborations; providing simple to use collaborations based upon advanced conferencing capabilities. The second generation Access Grid Toolkit provides a solid basis for more advanced collaborative applications to be developed. This tutorial will give an in depth overview of the Access Grid Toolkit version 2.0, highlighting improvements and new functionality.
The Access Grid (AG), known as the ensemble of resources that can be used to support human interaction
across the grid, consists of multimedia display, presentation and interactive environments, interfaces
to grid middleware, and interfaces to visualization environments. In this talk, focusing on the current
1.x version of AG, we will first introduce about the role and short history of it. Then, the hardware and
software setup for AG node will be followed with short guide on the operation of AG node. Later part
of talk will be allocated to introduce the on-going activities in both Korea and Austrailia to deploy and
improve the AG nodes.
B2: Overview of Digital Video Standards (Tutorial) (Room B 11:00-12:30, Jan 22)
Chair: Naohiro Shichijo <shichi@iii.u-tokyo.ac.jp> (Univ. of Tokyo, Japan)
During the last decade of 20th century,
television contents creation and broadcasting system
changed from analog to digital.
Almost all of the major television stations already adopted
digital cameras, VCRs, video effectors, editing systems etc.
The transition from special purpose audio/video network to
IP-based network is expected but yet to realize until now.
In this short course, we focus on digital video formats.
Since the formats of digital video are largely depend on those of analog.
The knowledge of analog video are required to understand them.
Firstly, a brief review of analog video signal formats are introduced
and its waveform and frame structure are described. Then digitizing
scheme of those analog signals are shown, which is known as composite
digital video formats.
The method to digitize luminance and chrominance signals separately to
improve color reproduction precision, component video digitizing
scheme, is presented next.
And multiplexing method of audio channels into those digital signals
are shown.
Finally, the various types of digital video tape recorders are
reviewed.
This is a beginners course on digital television systems.
Through this session, you will understand briefly what is performed
inside digital video systems, the digital video formats, audio/video
multiplexing method, kinds of digital VCRs and relation ship of those
standards with digital television broadcasting.
B3: QoS (Room B 14:00-15:30, Jan 22)
Chair: Lee Bu Sung, Francis <EBSLEE@ntu.edu.sg> (NTU, Singapore)
In this talk, we will cover our recent experience with on-demand QoS allocation and resource
management system (a.k.a. bandwidth broker) based on the IETF DiffServ QoS model.
In order to provide the end-to-end QoS, the implemented system consists of various component
modules: router resource analysis module, advance resource reservation module, enhanced
CAC module, QoS negotiation module, and SNMP-based monitoring module. The proposed
system provides user interfaces (by Java applet) that works both in interactive and automated
manners. We have performed the evaluation of the prototype implementation on KREONET/KOREN
testbed. The evaluation results including the 3D stereo video delivery test between KJIST and
KISTI will be reviewed to suggest several guidelines on the scalable design of QoS allocation
and resource management system.
In this talk, we present the experiences of integrating various advanced
technologies, viz. mobility support, quality of service (QoS) and security, in
an IPv6-based testbed. The purpose of the testbed project was to demonstrate
the feasibility of deploying emerging technologies in a progressive manner and
address the issues and conflicts arising from integrating different
technologies that were developed independently of one another. This enables us
to understand the problems that we will encounter when we deploy IPv6 with
these new emerging technologies.
Music delivery by Icecast has diffused. Icecast delivers music
to a lot of users by relay servers in real time, however,
the connection is cut off and the music pauses because of
retransmission control of TCP (Transmission Control Protocol)
in bad condition of network. TCP is the transport
layer protocol of communication between a server and relay
servers. We have proposed replacing TCP with UDP (User
Datagram Protocol) in communication between a server and
relay servers and used Reed Solomon code of FEC (Forward
Error Correction) to give guarantee of communication quality.
We have introduced and implemented the gateways into
the existing system preserving the existing applications, and
finally evaluated it in real network. As a result, it was shown
that the communication quality in the proposed system using
UDP and Reed Solomon code is better than the existing one
using TCP. In this way, we have reduced the pause and noise
maintaining the connection.
Keywords: Icecast, MP3, streaming, TCP, UDP, FEC, Reed Solomon
B4: VoIP (Room B 09:00-10:30, Jan 23)
The Internet provides a communications medium over which people
could interact. Videoconferencing and messaging are some of the new
applications that allow people real-time connectivity. The Session
Initiation Protocol (SIP) is a signaling protocol used for establishing
sessions in an IP network, over which other application layer protocols
could then be made to pass. This project aims to implement a SIP network
as described by RFC 3261 (June 2002), by developing UNIX-Based SIP
servers and user agents on Windows and Unix. The SIP server will provide
registration service, location service, proxy and redirect service. User
agents will run client applications with support for instant messaging,
audio and video.
SIP is a text-based protocol and can be compatible with various applications,
therefore it is more and more concerned with a VoIP and multimedia applications.
We convince that SIP is the next protocol for signaling instead of H.323.
So we have deployed the testbed, we can have tested lots of SIP based VoIP services such as
call forwarding, call blocking, etc. After the test, We divide SIP based VoIP
services into two types according to the call occurence. One type is a Call Processing
Based Service which a server can make a call by itself under a specific condition.
Now, CPL for the process of user call is on the standardization by IETF, but the Call
Reservation Based Service is not yes under discussion. So we propose CTM that can
make a call at specific time to process a reserved services and develop some SIP
based services using it. Besides CTM, we propose VoIP measurement system that
shows distinctly signaling time and media transmission time.
B5: Museum (Room B 11:00-12:30, Jan 23)
This research starts from the idea that through synchronized experiences between real and virtual museum, users can achieve thorough comprehension on the heritage . We investigated the virtual museum by constituting the concept of virtual museum with the reciprocal real museum. Based on the fundamental purpose and role of the museum, we suggested a model of a virtual museum composed by a virtual museum server and interface orbit. A user can contact the virtual museum services in the virtual museum server through various devices as interfaces. For the future works, we are developing “Virtual Namsan” integrated with technologies mentioned in this paper.
Always-on broadband network does not only enable sharing of high-quality 3D data, but also support networked virtual environments (NVEs) for remote users to collaboratively create something new and interesting contents. This talk introduces some building blocks to interactively digitize, visualize, and manipulate 3D objects in the NVEs. Our method uses a parametric approach to realize easy-to-deformable 3D shape representation. Some authoring application systems are shown to demonstrate how the proposed method easily and effectively help the 3D contents creation.
- The past 20years of museums and IT technology
- IT technology used at museums in future
- Digital archives of cultural assets
- Virtual museum in real museums
- Virtual museum in virtual space
B6: Security (Room B 14:00-15:30, Jan 23)
Chair: Sureswaran Ramadass <sures@cs.usm.my> (USM, Malaysia)
The fundamental problem with any network administration systems today is its ability to cope with the rising amount of virus intrusions. Currently available systems are only able to detect a virus after the network has been infected, therefore its non-real time. Depending on the malicious activities of the viruses, the detection will be carried out. Herewith, we are proposing a Real-Time Virus Detection system, which detects the arrival of virus intruders at the network layer rather than at the application layer. In this paper, we present an overview of the system design, which uses the iNetmon engine, Virus parser, Virus Matching Engine and alert mechanisms. Using the iNetmon engine, all packets traversing through the network nodes are captured; these packets are decoded and sent to Virus Matching Engine. Meanwhile, Virus parser will load the entire virus signature to memory. At the Virus Matching Engine, captured packet will be formatted to enhance matching speed. Then the formatted packet content will be scanned for virus information. Once the packet is known to contain virus or worm information, alert mechanism will alert the network administrator. Upon receiving this alert message, the administrator can now take necessary actions before the packet arrives at the destination.
Keywords: Virus, iNetmon Engine, Virus Parser, Virus Matching Engine
Today's information security market abounds with solutions that provide perimeter protection for a company's digital boundaries and they work well. Firewalls block unwanted traffic. Virtual Private Networks provide secure connections to partner sites. Intrusion Detection Systems warn about potential malicious attacks. However, mission critical data from customers, partners and employees needs to be permitted into the premises. It brings a wide variety of security issues that are complex to resolve and require massive integration across multiple vendor products. Our objective is to provide an application security platform to deliver a single protection source for all the application security needs of an enterprise. It aims at providing application security solutions protecting the data served to your clients, employees and business partners.
Safety and security ar two most important properties of a system. A safe system provides protection against error of trusted users, meanwhile a secure system protects against error introduced by untrusted users. In this situation, high requirement for rapid service creation have stimulated the development of programmable network infrastructure where end users or service providers can customize the properties of a network infrastructure while it continues to operate. The ability on customization of network infrastructure properties is called active networks. But the main concern of potential users of such systems is their reliability and most specifically their safety and security. As for this paper, the scope of the research is towards the safety issues involved in the active network architecture. A variety of proposals for exposing some control of network infrastructure have been made. Priority has been given on exposing this shared infrastructure to users must preserve some expectations of reliability while allowing the infrastructure to be multiplexed to derive the economics advantages of sharing. In this paper, attention has been given to one of this proposal, discuss further on the plus and minus point of it and also the future suggestion towards the better performance of the active networks.
Keywords: networking, active network, security aspect of active network, safety issues
B7: CRL Research Projects (Room B 16:00-18:00, Jan 23)
Chair: Kiyoshi Igarashi <igarashi@crl.go.jp> (CRL, Japan)
A program for comprehensive Arctic atmosphere observation by means of
advanced radio/optical remote-sensing instruments is being conducted
by Communications Research Laboratory (CRL), Japan, in cooperation with
University of Alaska Fairbanks (UAF). The targets includes understanding
various Arctic atmosphere phenomena (from ozone layer to aurora) related to
Global Environment/Changes. For effective use and display of the data
and for assistance in the data distribution, the System for ALaskan
Middle Atmosphere Observation Data Network (SALMON) is now being developed
and tested using high-speed links such as APAN, TransPAC, Abilene, etc.
The system have capabilities of quasi-real time data transfer to CRL
from Alaska, real time data displays on the world wide web
(http://salmon.crl.go.jp) as well as on-line data distribution
to data users and general public.
CRL's Yokosuka Radio Communications Research Center started a new
project named "Next Generation Mobile Network Project" in April 2002.
The target of this project is the development of new technologies to
enable seamless and secure integration of various wireless access
networks such as 3G and 4G cellular, wireless LAN, Bluetooth,
ultra-wideband (UWB), fixed wireless access (FWA) and intelligent
transport system (ITS). Three research groups of Wireless Applications,
Mobile Networking, and Wireless Access Groups and a Project Office for
promoting and planning the project are newly organized in the CRL
Yokosuka Center. A consortium for collaboration researches with
industries and universities is organized and a test bed for evaluating
new technologies is established in the Yokosuka Research Park (YRP).
In this presentation, I introduce the project and research activities
especially in mobile network group of the project.
Radio telescopes observing distant quasars are producing Gbps data
rate at their front-end AD converters. Interferometric observation
need the Gbps data transfer between telescopes and data correlation
extract high resolution scientific results from the radio stars.
VLBI(Very Long Baseline Interferometry) is the interferometric
observation between the separated telescopes.
In this decade former magnetic tape-based observations are move to
network observation depends on the network availability.
The network observation in not only bring quick results to researchers,
but also it enables quick follow up to sporadic astronomical events
which unable to observe before.
Japanese group had achieved first-ever Gbps-realtime VLBI in
June 2001. This experiment employed ATM-based data transfer system of NTT.
Since then, we carried out international experiments of Japanese-Finish
Gbps near-real time ftp observations and Japanese-US Mbps realtime IP
observation. Both were succeeded in October 2002, further growth
of back-bone network and international collaboration between telescope
sites and network researchers will enable us high sensitivity and
accurate observations to reach deep objects.
B8: P2P/NSCD (Network Storage and Content Distribution) (Room C 09:00-10:30, Jan 24)
Logistical Networking is an emerging technology for scalable deployment of
shared storage in the network. Intended as a adjunct to IP and other
conventional networking technologies, Logistical Networking enables the
network to support a number of functions, from asynchronous
store-and-forward communication to caching, content delivery, management of
massive files and flexible overlay routing. In this talk, I will give an
overview of the components of Logistical Networking, the currently worldwide
deployment including the Asia Pacific region, our current distribution of
end-user tools and new functionality that will be available soon. Finally,
I will briefly discuss our current research directions and development
plans.
Peer-to-Peer (P2P) file sharing applications based on Gnutella are
suffering with networking inefficiency due to their unstructured
nature. By leveraging the overlay network infrastructure,
We are currently attempting to improve the networking efficiency
of p2P file sharing applications. In this talk, we will review
existing overlay network approaches focusing on their topological
structure. Based on the review results, we will present our
initial thoughts on the scalable overlay network for P2P.
Copyright (C) 2003 APAN 2003 Conference in Fukuoka
$Id: B-abst.html,v 1.47 2003/01/23 05:31:59 kasahara Exp $
|