Research Topics in Ubiquitous Computing

1
Research Topics in Ubiquitous Computing

• Jason I. Hong
• jasonh at cs cmu edu
• http//www.cs.cmu.edu/jasonh/courses/ubicomp-f200
  4

2
Ubiquitous Computing is Coming
Devices in all form factors Sensors
everywhere Rich variety of inputs and outputs All
wirelessly connected Integrate computation,
communication, and sensing with physical world
3
Ubiquitous Computing is ComingDevices in All
Form Factors
4
Ubiquitous Computing is ComingDevices in All
Form Factors
5
Ubiquitous Computing is ComingSensors Everywhere
E911
Find a Place
Find Friend
6
Ubiquitous Computing is ComingSensors Everywhere

• RFIDs already in greater use than you may realize
• Wal-Mart mandate of Jan 2005

7
Ubiquitous Computing is Coming Rich Variety of
Inputs and Outputs
8
Ubiquitous Computing is Coming All Wirelessly
Connected

• Short-range wireless for consumers
• Bluetooth, HomeRF, 802.11
• Near Field Networks
• Personal Area Networks
• Transmit data thru the body

9
Just Where Are We Heading?

• Ubicomp field currently very disparate
• Systems, networking, HCI, Machine Learning,
• Different communities, different vocabulary,
  different papers
• Goals of this course
• Establish a common foundation for ubicomp
  research
• Help advance state of the art
• Difficult, but many opportunities for defining
  the field!

10
Who Am I?

• Undergrad at Georgia Tech, CS and Math
• Cyberguide, first mobile tour guide
• PhD in Computer Science from Berkeley
• Dissertation work on privacy and ubicomp systems
• New faculty at HCII
• Interests in ubicomp, focusing on privacy, rapid
  prototyping, end-user programming, and
  deployability issues
• Office at NSH 3613
• jasonh at cs cmu edu

11
Who Are You?

• Name
• Background
• PhD, MS, or undergrad?
• What year?
• Which unit? (ie CSD, HCII, CALD, RI, etc)
• Interests
• What kinds of research are you doing (if
  applicable)?
• What do you want to get out of this course?

12
Structure of this Course

• Research analysis synthesis
• Analysis understanding others work, the good
  and the bad
• Difficulty is that there are few provably correct
  answers
• Synthesis finding new ways to advance state of
  the art
• Literature survey
• Mini-projects
• Course project

13
Literature Survey

• Read, analyze, critique papers
• All research projects fail in some way
• Successful projects get some interesting results
  anyway
• Class preparation
• Reading papers is hard, especially at first
• Read before class
• Brief review of each paper (bring a printout)
• ½ page or less
• 2 most important things about that paper
• 1 major flaw (or ideas for extension /
  verification)
• Will see if can switch to Blackboard online
  discussions

14
Class Format

• Each person will do 10-20min presentation for a
  day
• Overview of readings
• Highlight what you think are interesting points
• Outline some questions for discussion
• Rest of time devoted to discussion
• What are the problems this paper addresses?
• How well does it address them? Realistic?
  Deployable?
• How to extend this research?
• Grad class, material may be controversial
• Ill do occasional presentations too

15
Overview of Course Topics

• Visions and Challenges of Ubiquitous Computing
• Context-Awareness
• Location sensing, uncertainty, prototyping tools
• Applications
• Home, workspaces, smart mobs, developing
  countries
• Privacy
• System Architectures
• Sensor networks, RFIDs

16
Mini-Projects

• Small projects to give you a flavor of ubicomp
• A few days of work
• Encouraged to work in groups
• Basis of mini-projects
• Topiary
• Place Lab

17
Mini-ProjectsTopiary

• Quickly create mockups of location-based apps
• Run these mockups with real users to get
  feedback

18
PlaceLab for Acquiring Location

• PlaceLab location via local database of WiFi
  Points
• Unique WiFi MAC Address -gt Latitude, Longitude
• Periodically update your local copy

• Works indoors and
• in urban canyons
• Works with encrypted nodes
• No special equipment
• Privacy-sensitive
• Rides the WiFi wave

19
PlaceLab
SF Bay Area 60000 Nodes (4 Megs)
20
PlaceLab
University of California Berkeley
Berkeley Campus 1000 Nodes
21
Course Project

• Do a small piece of real research
• Teams of 2-3
• Conference paper as deliverable
• Want best papers to be published in a real
  conference
• (With some extra work)
• Will put some project ideas online at website
• You can choose your own projects as well

22
Summary of Your Responsibilities

• Read papers before class
• ½ page summaries for each paper
• Signup to do a presentation
• Mini-projects
• Course project

23
Next Course

• Monday is Labor Day, no class until next Wed
• Readings
• The Computer for the Twenty-First Century, by
  Mark Weiser
• Synthetic Serendipity, by Vernor Vinge
• Mike Villa's World, by Harry Goldstein
• Only two summaries for next time
• One for Weisers paper
• One for Synthetic Serendipity and Mike Villas
  World
• Will see if I can get coffee, tea, and snacks

24
Important Questions to Think About

• Original vision of ubicomp written in 1991!
• Why dont we have it today? What would it take?
• How to manage this complexity?
• Things are barely interoperable today
• Can barely manage the flow of info today
• What are the core devices?
• Buy it at Home Depot, plug it in, and youre good
  to go!
• What are the core services of this world?
• Equivalent of Google, Yahoo, and Ebay?
• What are the programming abstractions?
• Is this really a world we want to live in?

25
The Future of Ubicomp?
26
Questions?

• Office at NSH 3613
• jasonh_at_cs.cmu.edu
• http//www.cs.cmu.edu/jasonh/courses/ubicomp-f200
  4

27
(No Transcript)
28
Bootstrapping Initial Apps Web Page Location
Auto-Fill

• Auto-fill location info on web sites

PageModification URL http//www.starbucks.com/ tx
tCity CityName txtState RegionCode txtZip ZIPCo
de
MapQuest
Starbucks
29
What does ubicomp enable? New Forms of
Entertainment
30
One Vision of Ubicomp

• We will reach a point where the combination of
  powerful processors, limitless data-storage
  capacity, ubiquitous sensor networks, and deeply
  embedded user interfaces will create a bond
  between human and machine so intimate that users
  may reasonably be considered superhumanly
  intelligent. - Vernor Vinge