English

应用科学学报 ›› 2014, Vol. 32 ›› Issue (2): 125-132.doi: 10.3969/j.issn.0255-8297.2014.02.003

• 通信工程 • 上一篇    下一篇

流水线型局部加权回归RFID 室内定位

张金艺1,2,3， 张晶晶1， 李若涵1， 徐德政2， 徐秦乐2   

  1. 1．上海大学特种光纤与光接入网省部共建教育部重点实验室，上海200072
    2．上海大学微电子研究与开发中心，上海200072
    3．上海大学教育部新型显示与系统应用重点实验室，上海200072
  • 收稿日期:2013-03-05 修回日期:2013-10-22 出版日期:2014-03-25 发布日期:2013-10-22
  • 作者简介:张金艺，研究员，研究方向：通信类SoC设计与无线传感器网络，E-mail: zhangjinyi@staff.shu.edu.cn
  • 基金资助:

    上海市教委重点学科资助项目基金（No.J50104）；上海市科委资助项目基金（No.08706201000, No.08700741000）

Pipelined RFID Indoor Positioning Based on Locally-Weighted Regression

ZHANG Jin-yi1,2,3, ZHANG Jing-jing1, LI Ruo-han1, XU De-zheng2, XU Qin-le2   

  1. 1. Key Laboratory of Special Fiber Optics and Optical Access Networks, Ministry of Education,Shanghai University, Shanghai 200072, China
    2. Microelectronic Research and Development Center, Shanghai University, Shanghai 200072, China
    3. Key Laboratory of Advanced Displays and System Application, Ministry of Education,Shanghai University, Shanghai 200072, China
  • Received:2013-03-05 Revised:2013-10-22 Online:2014-03-25 Published:2013-10-22

PDF

585

被引次数

6

摘要： 射频识别技术（radio frequency identification, RFID）以其非接触、非视距、低成本及高精度等优点成为室内定位技术的研究热点. 为了加强信号稳定性并提高实时性，该文用流水线方式接收到的包信息作为定位信号参数，针对室内环境对信号传播影响的复杂性，提出了流水线型局部加权回归定位算法，将室内环境对信号传播到各位置的影响融合进算法，以实现精确定位. 实验表明，对于室内定位，所提出的基于RFID 技术的流水线型局部加权回归定位算法相对于经典的LANDMARK 算法和VIRE 算法，定位精度分别提高56.56% 和36.73%. 在多目标的情况下，也可以实现实时精确的定位跟踪，具有良好的实用价值和应用前景.

关键词: 室内定位, 射频识别, 局部加权回归, 流水线, 收包率

Abstract: Radio frequency identification (RFID) is a hot research topic for indoor positioning because it is non-contact and non-line-of-sight with low-cost and high-precision. This paper proposes to use the pipelined packet reception rate as the positioning parameters to enhance stability and improve real-time performance.To deal with the indoor environmental impact on the signal propagation, the pipelined positioning algorithm uses locally weighted regression, which takes full advantage of the indoor environment information to achieve precise positioning. Experimental results show that, for indoor applications, the proposed algorithm increases the positioning accuracy by 56.56% compared with LANDMARK and by 36.73% compared with VIRE. The method can also obtain precise real-time tracking results for multiple targets, showing its practical value and wide range of applications.

Key words: indoor positioning, radio frequency identification (RFID), locally weighted regression, pipeline, packet reception rate

中图分类号: