1 / 18

Discovering Semantically Meaningful Places from Pervasive RF-Beacons

Discovering Semantically Meaningful Places from Pervasive RF-Beacons. Outline . Introduction Geometry-based vs Fingerprint-based Place Learning The PlaceSense Algorithm Experiment Data Collection Implementation Evaluation Metrics Results. Introduction . Place learning algorithm :

cameo
Download Presentation

Discovering Semantically Meaningful Places from Pervasive RF-Beacons

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. Discovering Semantically Meaningful Placesfrom Pervasive RF-Beacons

  2. Outline • Introduction • Geometry-based vsFingerprint-based Place Learning • The PlaceSense Algorithm • Experiment • Data Collection • Implementation • Evaluation Metrics • Results

  3. Introduction • Place learning algorithm : • Important to a individual user and carries a semantic meaning • Input : sequence of time-series sensor data • Output: sequence of tuples (date,entertime, leave time, place name)

  4. Geometry-based vs Fingerprint-based • Geometry-based(GPS): • Time-based clustering: compare the distance between the mean of the current cluster and the new measurement against the distance threshold • Fingerprint-based(WIFI): • BeaconPrint: using multiple scan windows to distinguish beacons, compares the fingerprint seen by the device to a list of place fingerprints learned by the system

  5. The PlaceSense Algorithm • continuously monitoring the radio beacons in the environment ( wifi AP or cell tower ID) around a mobile device.

  6. Entrance to a place: • Window size : W • Stable depth: s: 0~smax • smax × w

  7. Departure from a place: • representative beacon • responsive rate: rep • Tolerance depth: t: 0~tmax

  8. Buffering strategy • Rapidly detect place entry after quick transitions. • PlaceSense buffers overlapping data and starts entry determination in parallel

  9. Experiment • human participants to log any place they visited and stayed for more than five minutes

  10. Data Collection • Nokia N95 mobile phone • integrated GPS and built-in Wi-Fi. • Sampling rate : 0.1hz • Data collectors were asked to stay at a place for a predefined amount of time • ground-truth: each data collector was asked to keep a diary of the name and time they entered and left every place they stayed more than 5 minutes during the experiment

  11. Evaluation Metrics

  12. Implementation • time-based clustering algorithm • Beaconprint: • PlaceSense: smax=3, tmax=3

  13. Results

  14. Results

  15. Results

  16. conclusion • It uses response rate to select representative beacons and suppresses the influence of infrequent beacons • PlaceSenses accuracy gains are particularly noticeable in challenging radio environments where beacons are inconsistent and coarse • PlaceSense is accurate at discovering places visited for short durations

More Related