No abstract

No abstract

No abstract

Unreliability of the data streams generated by RFID readers is among the primary factors which limit the widespread adoption of the RFID technology. RFID data cleaning is, therefore, an essential task in the RFID middleware systems in order to reduce reading errors, and to allow these data streams to be used to make a correct interpretation and analysis of the physical world they are representing. In this paper we propose an adaptive sliding-window based approach called WSTD which is capable of efficiently coping with both environmental variation and tag dynamics. Our experimental results demonstrate the efficacy of the proposed approach.

Due to the high sensitivity of RFID tag-reader performance to the operating environment, RFID data streams generated are unreliable and contain a significant amount of missed readings. RFID data cleaning is therefore an essential task for successful deployment of RFID systems. One of the common techniques used by RFID middleware systems to compensate for the missed readings is the use of sliding-window filters. However, setting an optimum window size is non-trivial task especially in mobile tag environments. In this paper we present a new adaptive data cleaning scheme called WSTD based on some of the concepts proposed in SMURF but with an improved transition detection mechanism. WSTD uses the comparison of the two window subrange observations or estimated tag counts to detect when transitions occur within a window. In the mobile environment, our experimental results show that the WSTD scheme performs better than SMURF producing an improvement of about 30% less overall errors than that produced by SMURF.