基于MBF的RFID冗余数据清洗  

RFID Duplicates Data Elimination Based on MBF

作　　者：付文秀[1] 李星[1] 

机构地区：[1]北京交通大学电子信息工程学院,北京100044

出　　处：《铁道学报》2013年第7期85-89,共5页Journal of the China Railway Society

基　　金：国家高技术研究发展计划(863计划)(2006AA04A106)

摘　　要：RFID无线射频识别技术在铁路物流中的应用明显提高了管理效率。为了防止漏读数据,实际应用中通常一个区域会部署多个阅读器,这就导致数据的冗余。由于RFID数据是以流的形式快速、自动地产生,是一种动态数据集,很难一次以有限的内存清除冗余数据。本文提出基于矩阵型Bloom滤波器MBF(Matrix Bloom Fil-ter)的清洗方法TIMBF(Time Interval MBF)。因为基于MBF,它可以用来表示动态集合,并且不会产生消极错误,能以较小的内存获得很高的正确率。TIMBF保存了标签数据的读取时间,为了进一步减少内存空间的使用,提出了空间优化措施。实验结果证实,本文算法能以较小的内存有效地清除冗余数据。Application of radio frequency identification in railway logistics greatly improves management effi ciency. To prevent missing readings in application, multiple readers are deployed in one area. RFID data pro duce many duplicates. Because RFID data are generated in a streaming fashion as dynamic sets, it is difficult to remove duplicates in one pass with limited memory. In this paper, we proposed a method Time Interval MBF, which was based on MBF. MBF can get high correct rates with a small amount of memory. Due to Time Inter val MBF is based on Bloom filter, it does not produce false negative errors. TIMBF Saves the reading time of tag. To further reduce the memory space, we proposed space optimization. Experimental results show that our approaches can effectively remove duplicates in RFID data streams in one pass with a small amount of memory.

关 键 词：RFID 铁路物流 冗余数据清洗 MBF 动态集 

分 类 号：TN311[电子电信—物理电子学]