An Improved Algorithm for Reducing False and Duplicate Readings in RFID Data Stream Based on an Adaptive Data Cleaning Scheme
RFID technology is increasing rapidly and being successfully applied to various sectors such as supply chain management, manufacturing, retail, warehouses, wall-mart and health care applications. The data stream produced by RFID system is sometimes unreliable, inaccurate, low level. As a result, there is a heavy load of unreliable data which is useless for higher level application. Therefore, RFID data cleaning is the most important task for the successful deployment of RFID systems. Unreliable data appears in three forms of reading errors: false positive (unexpected), false negative (missed) and duplicate readings.