基于BP神经网络的RFID室内定位算法研究  

Research on RFID Indoor Localization Algorithm Based on BP Neural Network

作　　者：吴超[1] 张磊[1] 张琨[1] 

机构地区：[1]河北工业大学控制科学与工程学院,天津300130

出　　处：《计算机仿真》2015年第7期323-326,共4页Computer Simulation

基　　金：河北省高等学校科学技术研究指导项目(Z2012135)

摘　　要：针对室内物体或者人员使用主动式射频识别(RFID)技术问题,设计了适用于室内环境的定位算法。传统的RFID室内定位算法由于室内环境复杂多变,路径损耗系数很难准确估计,定位精度不高。为解决上述问题,提出一种采用BP神经网络的RFID室内定位算法,算法引入参考标签辅助定位,利用BP神经网络建立场强信号转化模型。在模型中输入接收信号强度值,输出路径损耗系数,网络模型提高了路径损耗系数的准确性,再利用距离-损耗模型实现精确定位,从而减小定位误差。与传统的RFID定位算法比较,仿真和实验结果表明,改进算法的定位精度有显著的提高,可用于仓库、监狱、超市等室内环境。For the probem that indoor objects or persons use active radio frequency identification technology,the paper presented an RFID indoor positioning algorithm based on BP neural network. The localization algorithm introduced reference label assisted positioning,and used BP neural network to establish the field strength signal conversion model. In the model,input received signal strength value,output the path loss coefficient,improved the accuracy of the path loss coefficient,and reused the distance-loss model to achieve precise positioning,thereby reducing the positioning error. Compared with traditional RFID localization algorithm,the simulation results show that the positioning accuracy of the method has been significantly improved,there are potential applications in warehouses,prisons,supermarkets and other indoor environments.

关 键 词：射频识别 神经网络 室内定位 信号强度 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]