Mobiscopes%20for%20Human%20Spaces

1
Mobiscopes for Human Spaces

• By Tarek Abdelzaher, Yaw Aanokwa, Peter Boda,
  Jeff Burke, Deborah Estrin, Leonidas Guiba, Aman
  Kansal, Samuel Madden, Jim Reich
• Presentation By
• Ankit Gupta

2
About the talk

• General Idea
• Why Mobiscopes?
• Classes of Mobiscopes
• Common Requirements
• Mobility and Sampling coordination
• Heterogeneity
• Privacy
• Networking Challenges
• Human Factors Social Implications
• Conclusion

3
General Idea

• Federation of distributed mobile sensors
• Why?
• Covering large areas can be challengeing
• Unavailability of wired power
• Expense of purchasing maintaining enough
  devices
• The paper focuses on the challenges and
  opportunities Mobiscopes pose in human spaces.

4
Classes of Mobiscopes

• Vehicular Mobiscopes
• For traffic and automotive monitoring
• Equipped vehicle senses various surrounding
  conditions
• Benefit
• Exploit oversampling provided by dense vehicle
  traffic
• Examples
• Inrix, EZCab, NavTeq, TeleAtlas etc.

5

• HandHeld Mobiscopes
• Could be useful for
• Monitoring health impact of exposure to highway
  toxins,
• Monitoring an individuals use of transportation
  systems,
• Gather real time information about civic hazards
  hotspots.

6
Common Requirements

• Data persistence must be assured
• Data access tends to be spatially correlated with
  the users location can change rapidly
• Human in the loop as an actuator, sensor,
  interpreter, or responder
• Sensors data to be shared by many public and
  private entities
• Trust, coordinated deployment and respect of
  userss privacy

7

• This all leads to
• General architecture and design guidelines for
  future Mobiscopes
• Component reuse and reduction in development
  costs
• Interoperability amongst future systems

8
Mobility and Sampling Coordination

• Performance depends on patterns of transporters
• Highly structured (Road traffic)
• Less structured (foot traffic)
• Sensor densities
• Sensing devices availability can depend on user
  behavior or device characteristics

9
(No Transcript)
10

• Application Adaptation
• Must adapt to networks available communication
  characteristics
• Could buffer data when connectivity unavailable
• Actuated Mobility
• Task some or all nodes to visit a specific
  location to collect information on demand
• Task actuators to visit some areas either one at
  a time or as part of a circuit

11

• Opportunistic connectivity
• Building low-level network protocols to quickly
  identify and associate with nearby node (or
  networks)
• Routing algorithms to deliver data through such
  opportunistic connections
• Prioritization
• Buffered data to be prioritized
• Prioritization to avoid wasting valuable
  bandwidth when different nodes cover overlapping
  geographic areas

12
Challenges and opportunities of heterogeneity

• Mobiscopes take on various topologies
  structures
• Federate devices with different capabilities
• Draw together components with varying levels of
  trust credibility
• Benefits
• Immune to weaknesses of sensing modalities
• Robust against defective, missing or malicious
  data sources

13
(No Transcript)
14

• Heterogeneity of Ownership
• Individually owned devices
• Owners might not be trustworthy
• Might not maintain their devices in good
  condition
• Data Resolution Types
• Derive maintain metrics at multiple resolutions
• Simple interpolations (smoothly varying,
  temperature)
• Complex models (faster varying or sparse data)

15

• Robustness
• Model driven approaches like Kalman filters
  Particle filters adapt well to irregular sampling

16
Tackling data Privacy

• Peoples ability to control information flow
  about themselves
• Definition
• Inability to publicly associate data with sources
  could lead to los of context
• Revealing too much context can potentially thwart
  anonymity, violating privacy requirements

17

• Local Processing
• Putting the selectivity and filtering
  capabilities on the end-user
• Verification
• Important to develop systems where users can
  verify datas correctness without violating the
  sources privacy
• Proper incentives to promote successful
  participation, prevent abusive access with the
  purpose of Gaming the system

18

• Privacy preserving data mining
• User isnt willing to share his or her data, but
  might be interested in the result of aggregation
  over the target community
• Could use additive random noise to perturb data
  withour affecting the statistics to be collected

19
Networking Challenges

• Shifts the networks main utility from data
  communication to information filtering
• Need for network storage as a key service because
  aggregation and filtering both imply a need to
  buffer

20
Human Factors and Social implications

• Considering broader policy precedents in
  information privacy
• Extending popular education on ITs new
  observation capabilities
• Facilitating individuals participation
• Helping users understand audit their own data
  uploads

21

• User Interfaces
• Missing from traditional embedded systems
• Opportunity for ambient and explicit feedback to
  the user
• Help users configure their sensing participation
• Provide feedback on operational status

22
Conclusion

• Much research still needs to be done
• Much work still needs to be done on
• Platforms APIs that offer efficient, robust,
  private secure networking sensory data
  collection in the face of heterogeneous
  connectivity and mobility

23
Questions

• ???