Dynamic Data Delivery to Mobile Users

1
Dynamic Data Delivery to Mobile Users
2
Outline

• Introduction
• Related Work
• Dynamic Channel Model
• Ubidata
• Methodology
• Application
• Conclusion Future Work

3
Introduction

• Prefetch anticipate mobile users
  needs for information
• This paper extension above concept for
  information channels,capturing mobility-related
  attributes
• Dynamic channel
• Adding adaptation capability to channel

4
Related Work

• Information Dissemination Models
• Based on the push concept
• - Smart pull (periodic pull)
• - True push (aperiodic push)
• Two cost incurred in moblie environments
• - access cost
• - communication cost
• Dynamic view is a method to reduce costs.
• This paper reduce costs by filtering information
  relevant to the mobile client.

5
Related Work

• Markup Languages for Information Dissemination
• Push systems need a common format to describe
  what information is sent to receivers.
• XML
• CDF (Channel Definition Format)
• Extend CDF to support mobile computing

6
lt?XML version"1.0"?gt ltCHANNEL
HREF"http//sample.microsoft.com/Channel/homepage
.htm" BASE"http//sample.microsoft.com/Channel
/"gt ltTITLEgtSample "Color" Channellt/TITLEgt
ltABSTRACTgtThe sample color channel contains
a red, green, and blue page for
viewing.lt/ABSTRACTgt ltLOGO HREF"logo_big.gif"
STYLE"IMAGE-WIDE"/gt ltSCHEDULE
STARTDATE"1997-09-23"gt ltINTERVALTIME DAY"1"
/gt ltEARLIESTTIME HOUR"2" /gt ltLATESTTIME
HOUR"6" /gt lt/SCHEDULEgt ltITEM
HREF"page1.htm"gt ltLOGO HREF"red.gif"
STYLE"ICON"/gt ltTITLEgtThe Red Pagelt/TITLEgt
ltABSTRACTgtThis is the abstract description for
the red page.lt/ABSTRACTgt lt/ITEMgt ltITEM
HREF"scrnsave.htm"gt ltUSAGE
VALUE"ScreenSaver"gtlt/USAGEgt lt/ITEMgt lt/CHANNELgt
7
Dynamic Channel Model

• The main classes of a channel are transmitter and
  tunner
• Transmitter (zero or more channels)
• Channel (items,subscriptions)
• Item (view,region,schedule)
• Notification
• Tuner
• Subscription

8
(No Transcript)
9
Ubidata

• Provide three functionalities
• Information dissemination to reachable
  subscribers
• Noticfication of changes to subscribed data
• Definition of individual schedules for each of
  the subscribed items

10
(No Transcript)
11
Methodology

• Information identification characterization
• Extend CDF to capture mobility related
  information
• Both Publisher subscriber take the CDF as input

12
Application
13
Application
14
Conclusion Future Work

• Developed a new editing tool for channels a
  generator of D-CDF
• Make it possible to disseminate information to
  smaller mobile units
• Ubidata in WAP