Institutional Repository
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Adaptive cleaning for RFID data streams

Shawn R. Jeffery, Garofalakis Minos, Franklin Michael J.

Full record


URI: http://purl.tuc.gr/dl/dias/CC96CE04-CBF2-4254-99E9-60E54877435C
Year 2006
Type of Item Conference Full Paper
License
Details
Bibliographic Citation S. R. Jeffery, M. Garofalakis and M. J. Franklin, "Adaptive cleaning for RFID data streams", in 32nd International Conference on Very Large Data Bases, 2006, pp 163-174.
Appears in Collections

Summary

To compensate for the inherent unreliability of RFID data streams,most RFID middleware systems employ a “smoothing filter”, asliding-window aggregate that interpolates for lost readings. In thispaper, we propose SMURF, the first declarative, adaptive smoothingfilter for RFID data cleaning. SMURF models the unreliabilityof RFID readings by viewing RFID streams as a statistical sampleof tags in the physical world, and exploits techniques grounded insampling theory to drive its cleaning processes. Through the use oftools such as binomial sampling and π-estimators, SMURF continuouslyadapts the smoothing window size in a principled manner toprovide accurate RFID data to applications.

Services

Statistics