Networking of Sensor Systems

1
Networking of Sensor Systems

• Guru Parulkar
• National Science Foundation
• Gparulka_at_nsf.gov

2

• While the last 50 years have been dominated by a
  march to ever more complex computers, the next
  few decades will see the rise of simple sensors
  -- by the billions. Business Week

3
Agenda

• Sensors and sensor networks opportunity
• Range of activities in progress
• Motivation for the focus area
• Focus of the focus area
• Partnership between research community and NSF
• Accelerate progress and realize full potential

4
Sensors and sensor networks represent a great
opportunity
5
Disruptive Technologies

• Digital sensors and actuators
• Very inexpensive and can be integrated into
  silicon
• Wireless
• Low power inexpensive RF
• Silicon integration
• Sensor, DSP, CPU, FPGA, wireless, actuators
• Huge embedded software on a chip/device
• Moving to Smart Dust

UC Berkeley Motes, Crossbow Sensors
6
Applications in All Aspects of Life
Intel Presentation
7
Sensors and sensor networks have taken off
8
Range of Activities in Progress
R D Results
Federal Funding
ResearchCommunity
StartupCorporate
9
Range of Activities in Progress

• NSF
• STC (CENS), ERC (CASA)
• Sensor and Sensor Networks
• EHS
• ITR numerous other grants
• DARPA
• SenseIT
• NEST
• DHS/HSARPA
• Others

R D Results
Federal Funding
ResearchCommunity
StartupCorporate
10
Range of Activities in Progress

• Increase in proposals
• 1000 proposals to NSFs Sensor and Sensor
  Networks
• New graduate courses
• PhD theses
• ACM SenSys
• ACM Journal
• Several workshops
• Conference themes

R D Results
Federal Funding
ResearchCommunity
StartupCorporate
11
Range of Activities in Progress

• Platforms
• Dust, Motes, Micro-servers
• TinyOS, TinyDB
• Algorithmic solutions
• MAC, routing, localization, time synch, tracking,
  querying
• Applications
• Simple to complex
• Testbeds
• Simple to elaborate

R D Results
Federal Funding
ResearchCommunity
StartupCorporate
12
Range of Activities in Progress
R D Results
Federal Funding

• Startups
• Dust Inc, Crossbow, Ember, AWS, Digital Sun, and
  more
• Corporate Interest
• Intel, TI, Agilent, IBM, HP, Lucent, Motorola,
  and more

StartupCorporate
ResearchCommunity
13
Why this focus area? Why more funding?
14
Top Five Reasons
5. Sensor network has potential to transform our
society
4. Existing programs too focused or too broad
3. Focus on networking will have big payoffs
2. Enable common architecture, interfaces,
reusable systems
1. Enable sensor networks to be commodity easy to
deploy tech
15
Sensor Networking Today Too Vertically Integrated

• Natural for an emerging area
• Stove pipe approach cannot work for wider
  deployment

16
Need Plug and Play Sensor Network Substrate

• Conceptual framework

17
Commodity Solution Easy to Deploy
18
Focus of the Focus Area?
19
Focus Drilling Down

• Monitoring Managing Spaces and Things

Applications
NetworkProgramming
HW/SWSystems
ProtocolsAlgorithms
PrivacySecurity
NetworkArchitecture
Store
Comm.
uRobots actuate
MEMS sensing
Proc
Power
technology
Miniature Connections to Physical World
20
Network Architecture (John Doyle Observations)
Variety of Ligands Receptors
Bowties and hourglasses are universal
architectures (or protocols suites)
Intermediates
Variety of responses
Nutrients
Products
21
The Internet Hourglass
Everything on IP
IP
22
Hourglass/Bowtie for Sensor Networks?

• Likely the most important innovationas Internet
  architecture was for data networking

23
Protocols/Algorithms Challenge

• Given
• New types of devices with differing capabilities
• A variety of wireless and some wired links
• Different connectivity and failure modes
• Applications with very different requirements
• Need to invent new protocols, algorithms, their
  implementations

24
System Design Challenge

• Challenging space of trade-offs
• Energy vs duty cycle
• Failure rates and redundancy
• Computing vs communications
• RF power vs range
• Time
• Space
• Need new design and verification models,
  methodologies and tools to help system design
• Need new hardware and software systems
• Sensor node, OS, programming environment,
  in-network processing

25
Security and Privacy Challenge

• Very different and challenging
• Physical environment cannot be protected
• Traditional firewall key distribution
  approaches do not work
• A compromised SES or application can
  easilydisrupt, carefully alter, or snoop
  operation
• Damage would be very serious
• Need new thinking and solutions
• Implications on hardware, OS, networked
  programming tools
• Have just barely started

26
Network Programming Challenge

• Need to program a group of potentially 1000s of
  sensor systems with a set of unique constraints
• Need new programming models and tools
• equivalent of socket interface
• Increase ease of programming and reusability
• New abstractions and automatic mapping on an
  array
• Assure privacy and security
• In-network computing an integral part

27
Potential Breakthrough Implications

• New radio technology
• E.g., UWB low power, high data rate, good range
• Mainstream applications
• Sub 100 nodes single hop networks
• Fuel cell based energy source
• E.g., Hydrogen fuel cell energy is no more
  constraint
• Low power chip designs
• Others that can change fundamental trade-offs?

28
Recap on Focus

• Monitoring Managing Spaces and Things

Applications
NetworkProgramming
HW/SWSystems
ProtocolsAlgorithms
PrivacySecurity
NetworkArchitecture
Store
Comm.
uRobots actuate
MEMS sensing
Proc
Power
technology
Miniature Connections to Physical World
29
Focus Area Approach
Accelerate Progress Realize Full Potential
Close Coupling Required
30
Expected Results of Program

• Foundations research
• Theoretical, algorithms and systems
• Hourglass/bowtie?
• Systems
• Several networkable sensor platforms range of
  capabilities
• OS and protocols stacks
• Network programming environments, in-network data
  processing
• Many local testbeds with applications
• Education and Training
• New graduate and undergraduate courses with
  experimental projects
• Many graduate students with hands-on experience
• Identifiable distinct community with members from
  
• sensors, embedded hardware, os, networking,
  middleware, applications

31
Out of Focus

• Focus naturally means
• Emphasize and invest in select topics and
  approaches
• De-emphasize other topics
• De-emphasized topics may be important for other
  programs

32
Topics Outside The Scope

• Sensor development in isolation
• New applications by themselves
• A whole range of embedded systems
• System development with
• No new capabilities
• No research content
• Modeling and algorithmic efforts
• Not aimed at real systems
• Theoretical work
• Not well motivated with its real relevance
• If in doubt please ask or send email

33
A Proposal that is NOT a Match

• represents tremendous cost to all parties and
  reduces effectiveness of the entire system
• Costs include precious time and efforts of
• PIs
• Graduate students and staff
• Departmental administrative staff
• Research office staff
• NSF administrative staff
• NSF program director(s) division director
• Reviewers
• Reduces pool of reviewers

34
Partnership Research Community and NSF
35
Need Partnership at Every Step of The Process
Focus AreaCommitment

• Send your best ideas consistent with focus area
• Save others for appropriate solicitation
• DONT submit same proposal to multiple programs
• NSF committed to providing all help

FundingDecisions
36
Need Partnership at Every Step of The Process
Focus AreaCommitment
InformationalMeeting
ProposalSubmission
Review Panels

• Suggest strong panelists
• Volunteer to be a panelist
• NSF committed to running objectiveand thorough
  review process

FundingDecisions
CommunityBuilding
ExecuteResearch
HighlightResults
37
Need Partnership at Every Step of The Process
Focus AreaCommitment
InformationalMeeting
ProposalSubmission
Review Panels

• Committed to funding best in support of focus
  area goals
• Due consideration to high risk high reward prop
• Committed to be responsive and timely

FundingDecisions
CommunityBuilding
ExecuteResearch
HighlightResults
38
Need Partnership at Every Step of The Process
Focus AreaCommitment
InformationalMeeting
ProposalSubmission
Review Panels

• NSF funded workshops and PI meetings
• ACM/IEEE conferences, workshops, journals

FundingDecisions
CommunityBuilding
ExecuteResearch
HighlightResults
39
Type of Projects

• Depends on the proposed project
• Single investigator two grad students
  equipment travel
• to
• Multi-investigator multi-disciplinary several
  grad students equipment travel system
  building support
• 150k to 1M per year
• Typically three years
• Exact distribution will depend on submitted
  proposals
• 80 funding likely to larger projects (fewer in
  number)
• 20 funding likely to smaller projects (more in
  number)
• Testbed projects, curriculum development welcome
• Annual PI meeting, two workshops per year

40
Intels Equipment Support

• Intel to provide sensor equipment to successful
  projects
• Imotes, gateways/micro-servers, PCs
• Equipment spec can be obtained from Lakshman
• PIs are not required to use Intel equipment
• Other vendors welcome to contribute equipment
• Need open platforms

41

• Acknowledgements

42
Where We Want to Get
43
How to Get There
Applications
NetworkProgramming
HW/SWSystems
ReusableSystems Science
ProtocolsAlgorithms
PrivacySecurity
NetworkArchitecture
FoundationsResearch
ExperimentalSystems
Store
Comm.
uRobots actuate
MEMS sensing
Proc
Power
technology
44
Closing Thoughts

• Sensors and sensor networks represent a great
  opportunity
• With focus and shared vision we can accelerate
  progress and realize the full potential

45
Questions/Comments?

• Goals of the focus area?
• Target topics and type of research?
• Type of projects?