1 / 19

Adaptive Data Collection in Environmental Sensor Networks

Adaptive Data Collection in Environmental Sensor Networks. Jayant Gupchup gupchup@jhu.edu. Cub Hill Network. Question. Downloading data from everyone is expensive Radio usage consumes power Most of the time nothing interesting is happening Data is strongly correlated. Proposal.

fadey
Download Presentation

Adaptive Data Collection in Environmental Sensor Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Adaptive Data CollectioninEnvironmental Sensor Networks Jayant Gupchup gupchup@jhu.edu

  2. Cub Hill Network

  3. Question • Downloading data from everyone is expensive • Radio usage consumes power • Most of the time nothing interesting is happening • Data is strongly correlated

  4. Proposal • Download data from a representative set that captures the variation in data • Minimizes communication costs • Extends the lifetime of the network • Reconstruct data using a model and the representative sensors

  5. Challenges • How do you pick informative locations • How many do we need? • How stable is the working set? • How do you reconstruct data for the “unobserved” locations • How do we estimate/measure the reconstruction error

  6. Example

  7. Information Theoretic Approach

  8. How many locations ?

  9. Questions?

  10. Data

  11. Reconstruction using PCA basis

  12. Working set frequency

  13. Number of location

  14. Smarter way? • Information gain criteria • A represents working set of locations • V\A represents locations NOT selected

  15. Compare with Random

  16. Scree plot (median)

  17. Scree Plot (95th percentile)

More Related