基于电磁检测技术的金属标签信息提取关键技术研究

The perception and collection of the information of things plays an important role in the Internet of Things. And identification techniques have been researched for decades and bar code, RFID and magnetic stripe have been applied in many areas successfully. However under adverse circumstances, such as humidity, high temperature, pollution and so on, neither of them can works for a long time. Besides that, they all really cost and damage easily. Metal Tag has the advantages that it can work stably and be integrated as a part of goods by engraving symbols on the surface of metal products and tags, which consist of notch arranged in 1- or 2-D space. This subject focuses on mathematic models of symbols employed by Metal Tag and processing techniques, including both point and distributed mathematic models of symbols containing electromagnetic characteristics, the analysis of the uniqueness of information representation in electromagnetic inverse problem and mapping symbol identification problem to estimating the number and DOA's of source in array processing. In the case of near-field full coherent signals impinging on small-scale arrays, we propose a new array processing method. Meanwhile to improve resolution in DOA estimation and calibrate array position errors, we give the solutions. Through this subject, we will provide promising experience on metal tags automatic identification both in theory and practice. At the same time, the key techniques employed here can be applied to non-destructive inspection, automatic recognition and so on.

物品信息的感知与采集是物联网体系的重要组成部分，目前几种主要的物品识别技术（条码、RFID、磁条）存在着容易损毁、相对价格较高、不能在潮湿、高温、污染等环境下长期使用的缺点。金属标签系统在金属物品表面或金属标签表面刻制图形，通过空间排列构成信息标识系统，具有稳定性好、与物品一体化的优点。本课题研究了基于电磁感应检测技术的金属标签系统数学模型和数据处理方法。核心内容包括建立金属符号电磁特性数学模型，包括点源模型和空间分布式模型，以及电磁逆问题中信息表示的唯一性分析。利用阵列信号处理技术，将符号的识别问题转变为阵列信号问题中源数和方位估计的问题。提出了近场源全相干信号小尺度阵列处理方法，对提高方位分辨率技术进行了研究，给出了相关的解决方法，并研究了阵元误差的校正方法。通过本课题的研究，可以为金属标签自动识别提供可参考的理论和方法，同时其关键技术也可应用到无损检测，自动识别等领域。

为适应物联网在工业、军事等恶劣环境下对信息感知的要求，本课题提出一种金属条码标签方案。该方案将普通条码刻蚀在金属导体表面，检测时将激励线圈靠近金属导体表面，在其表面层中感生出电涡流，通过检测线圈阻抗变化或者金属近表面空间中的磁场分布，实现对金属条码的识别。.本课题致力于研究基于电涡流检测技术的金属条码设计及识别和基于非晶态合金微丝的空间磁场测量技术。.在电涡流检测部分，课题组通过并矢格林函数积分方程法给出金属条码对检测线圈阻抗变化的影响；通过薄膜场理论给出金属条码对条码近表面空间的磁场扰动。通过仿真分析，定性给出金属材质、激励频率、条码密度、单元尺寸等参数的选择。根据仿真所得参数，课题组设计了一套探头采用激励、检测线圈分置形式的金属条码识别系统。.在非晶态合金微丝巨磁阻抗研究部分，课题组研究了偏置磁场对非晶态合金微丝磁导率张量的影响；通过OOMMF微磁学仿真软件模拟了外磁场作用下非晶态合金微丝内部的磁化分布；通过有限差分方法模拟了小长径比非晶态合金微丝的退磁场分布；并设计了一款可用于测量非晶态合金微丝在不同磁场偏置条件下对角阻抗和非对角阻抗的阻抗分析仪。