西安电子科技大学学报

• 研究论文 • 上一篇    下一篇

一种利用无线信号的被动式多目标定位系统

邢天璋;谢彬彬;杨康;汤战勇;房鼎益   

  1. (西北大学 信息科学与技术学院,陕西 西安 710127)
  • 收稿日期:2017-03-19 出版日期:2018-02-20 发布日期:2018-03-23
  • 作者简介:邢天璋(1981-),男,高级工程师,E-mail: xtz@nwu.edu.cn
  • 基金资助:

    国家自然科学基金资助项目(61602381, 61672428, 61672427); 陕西省国际合作资助项目(2015KW-003)

Passive localization for multi-targets via wireless signal processing

XING Tianzhang;XIE Binbin;YANG Kang;TANG Zhanyong;FANG Dingyi   

  1. (School of Information and Science and Technology, Northwest Univ., Xi'an 710127, China)
  • Received:2017-03-19 Online:2018-02-20 Published:2018-03-23

摘要:

针对真实环境中的多径分布及非视距路径的多目标定位问题,基于指纹库匹配的思想,在无源射频识别标签环境中,提出一种被动式多目标定位系统.重点针对多目标的计数问题进行了深入的分析,通过无源标签边缘检测,将标签间距离与目标个数建立函数关系,并对阅读器碰撞现象进行分析,从而确保定位数据的有效性与目标统计的精确性; 其次,探索了目标相对于标签位置的相异性,并利用标签位置作为目标定位的约束条件,提升定位精度.在真实环境中对该系统予以测试,实验数据表明,该系统在目标个数统计与定位精度两个方面有效且均能满足设计要求,并较现有方法有明显提高.

关键词: 目标定位, 指纹匹配, 目标计数, 射频识别

Abstract:

Passive localization plays a critical role in the Internet industry. However, current methods only analyze and calculate the signal distortion of the target which generates in the Line of Sight without considering the Non Line of Sight and multipath effects. To address the problem, we present the multi-targets location system via the RFID base on the fingerprint matching. First, this paper uses the passive tag edge detection to build the function relation between the target distance and the target number, and analyzes the reader collision phenomenon to address multi-targets counting, further ensuring the effectiveness of location data and the accuracy of target statistics; Second, this paper improves localization accuracy by exploring the arteritis of the target relative to the tag position and using the tag position as the constraints. We conduct extensive experiments in 2 scenarios for 3 areas. Compared with other approaches, the system we proposed improves the target counting and location accuracy, scalability and robustness under various scenarios.

Key words: localization, fingerprint, target counting, radio frequency identification devices