基于RBF神经网络的RIFD多标签识读距离预测研究  

RIFD Multi-tag Reading Distance Prediction Based onRBFNeural Network

作　　者：王倩文 李琳[1] 刘振鲁 WANG Qian-Wen;LI Lin;LIU Zhen-Lu(College of Physics,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学物理学院,南京210016

出　　处：《电光系统》2022年第2期22-27,共6页Electronic and Electro-optical Systems

摘　　要：射频识别(RFID)作为一种非接触式的自动识别技术,因具有识读距离远,存储容量大等优点得到快速发展,广泛应用于智能交通、物流货物运输、智能畜牧等领域。多标签识读距离作为检验RFID识读性能的标准之一,研究多标签位置之间的关系对识读性能的提升具有重要的研究价值。文章围绕基于神经网络的RFID多标签识读距离预测开展研究,提出采用基于深度学习的径向基函数(RBF)神经网络预测识读距离的方法,RBF神经网络对通过RFID半物理仿真验证平台实验采集得到的数据进行训练学习,从而实现根据多标签三维坐标预测系统的识读距离。As a non-contact automatic identification technology,RFID(Radio Frequency Identification)is wide-ly used in intelligent transportation,logistics and goods identification,intelligent livestock and other fields with its long reading distance,large storage capacity and other advantages of rapid development.Multi-tag reader's reading distance is one of the criteria for testing RFID s reading performance,and studying the relationship be-tween the Multi-tag's location has important research value to promoting the reading performance.This paper fo-cuses on the prediction of RFID multi-tag readers reading distance and proposes a method of RBF neural net-work based on deep learning to predict the reading distance.The RBF neural network trains data which is ac-quired by RFID semi-physical simulation verification plaform and predicts the reading distance by the multi-tag 3D coordinate system.

关 键 词：RFID RBF神经网络 多标签 预测模型 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]