RFID数据流过滤算法研究

摘要

RFID在实时识别、定位、跟踪以及物理对象监控等方面有着广泛的应用前景.为了实现上述应用的目标,RFID数据必须经过采集、过滤等处理过程.采集的原始RFID数据包括噪音数据和冗余数据,只有将这些噪音数据和冗余数据过滤后,才能应用；另外,保证正确的RFID标签数据采集顺序,对于一些应用也非常关键.针对这些问题,提出了基本噪音过滤算法和基于hashtable的有序噪音过滤算法,以及基本冗余过滤算法和基于hashtable的冗余过滤算法,并通过仿真实验验证了该方法的有效性.%RF1D holds a wide ranges of pervasive computing applications of real-time identifying, locating,tracking and monitoring physical objects. To achieve these goals,RFID data contains false readings and duplicates has to be collected,filtered and transferred into semantic application data. Such data cannot be used directly unless they are filtered and cleaned. Meanwhile, the order perservation of RFID tag observations are critical for many applications. In the paper,propose serveral efficient methods to filter RFID data,including both noise removal and duplicate elimination. The performance study demonstrates the efficiency of the methods.