基于Bloom filter的RFID中间件数据过滤算法研究  

Research of RFID middleware data filtering algorithm based on Bloom filter

作　　者：冯锋[1] 吴杰[1] 

机构地区：[1]宁夏大学数学计算机学院,银川750021

出　　处：《计算机应用研究》2015年第5期1441-1444,共4页Application Research of Computers

基　　金：宁夏自然科学基金重点资助项目(NZ13004)

摘　　要：有效地减少RFID系统中冗余阅读器或天线采集到的大量重复数据,可以降低系统能耗和提高处理效率。经研究,提出采用改进的布隆过滤器(Bloom filter)对RFID采集数据进行去重过滤,并运用到中间件系统中。改进的Bloom filter主要将两个标准的Bloom filter组成二维并行Bloom filter,对RFID采集数据所包含的两个属性值tag ID和reader ID进行并行过滤。经实验可见,标准Bloom filter与哈希过滤(hash filter)相比具有明显的优势,对其改进后,采用二维并行Bloom filter在误判率、吞吐率和存储空间上具有更高的系统性能。In order to achieve the goal of reducing the energy consumption of system and improving the processing efficiency needs to reduce the duplicated data collected by redundant readers or antenna effectively in RFID system. Through the research,this paper put forward an improved Bloom filter to filter the duplicate data which was gathered by the RFID,and applied it to the middleware system. The improved Bloom filter primarily composed of two standard Bloom filters as a two-dimensional parallel Bloom filter. It parallel filtered the two attributes： tag ID and reader ID,which was gathered by RFID. The experimental result shows that standard Bloom filter offers significant advantages over hash filter,and the improved two-dimensions parallel Bloom filter has higher system performance on the misjudgment rate,throughput and storage space.

关 键 词：无线射频识别 中间件 布隆过滤器 哈希过滤 二维并行 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]