A data filtering strategy using cluster architecture in radio frequency identification system

Radio Frequency Identification (RFID) readers use radio waves to interact with the tagged objects and continuously transfer the reading data to the middleware. Because of the high error reading rate, RFID readers generate a large amount of duplicate and inaccurate data. In this paper, we propose a filtering model to correct and clean RFID data generated by RFID readers. We group the readers in an RFID system into several clusters according to their deployed locations. In every cluster, the cluster header uses the readings generated by the readers in the cluster and other clusters to assist the data cleaning processing. Our filtering model can not only increase the accurate rate of the reading but also share the heavy work load of the middleware. The experimental results show that our approach is efficient and accurate.