【作者】 李庆亮；

【导师】 施文康；

【作者基本信息】 上海交通大学 ， 精密仪器与机械， 2008， 博士

【摘要】 基于声表面波(SAW)技术的射频标签具有纯无源、阅读距离远、环境适应性强等特点,正在逐步受到大家的关注。本文对基于声表面波的射频标签系统和阅读器进行了理论研究,着重在提高标签识别率、多标签辨识和提高检测精度等方面。并在此基础上设计制作了一套完整的SAW标签系统。主要研究内容如下:首先,介绍了SAW标签系统的远场工作条件,分析了阅读器天线和声表面波标签天线的链路模型。并对SAW标签设计相关的SAW技术进行了论述。其次,讨论了SAW标签的各种编码方法及其特点;分析了反射栅多次反射的影响;讨论了SAW标签材料的选择。第三,在射频识别系统空间传播技术和信号检测原理分析的基础上,对SAW标签阅读器系统设计做了理论分析。阐述了时域采样和频域采样时的不同检测原理和系统组成及各自工作特点,据此确定本文采用时域采样的阅读器设计方案。着重分析了时域采样下的检测理论方法,包括振幅检测、相位检测和相位估计方法。指出在大信噪比时,采用振幅检测能够获得更高的检测概率;而在小信噪比时,相位检测则能够更好地利用有用信息。同时,针对振幅检测和相位检测在不同信噪比下的各自检测特点,采用基于检测距离的单标签融合检测方法来提高阅读识别率。第四,由于SAW标签是纯无源标签,不具有任何数据处理能力,因此多目标识别是SAW标签系统的一个难点。本文分析并指出了各种防冲突方法的优点和存在的不足。在此基础上,采用基于Walsh域的匹配门限滤波方法来对多目标进行辨识。并采用滤波门限的方法保证了运算速度。第五,反射脉冲的峰值和位置是标签解码的两个主要参数。本文采用基于最小均方误差的权重数据融合方法来估计标签的脉冲响应的幅度和位置信息。第六,设计了一套完整的SAW射频识别系统,包括SAW标签和阅读器。SAW标签是按照本文提出的多目标辨识方法筛选并制作。阅读器包括射频发射机、接收机、高速数据采集、数据信号处理等部分,实现了对标签信息的幅度检测和相位检测。并提出了提高阅读器射频通道隔离度的改进措施,在系统工作频段上,将阅读器查询信号的隔离度提高到60dB。最后试验测试了SAW标签辨识系统,其阅读距离达到4.5米以上,验证了基于检测距离来提高单标签识别率的融合方法,并测试了多目标标签的识别能力。本文的主要创新点:1.提出了基于检测距离的融合检测方法来提高单标签阅读识别率。2.提出了基于Walsh域的多目标辨识方法。并进一步提出采用滤波门限的方法保证了运算速度。3.提出基于匹配追踪-小波包原子分解估计的最小均方误差权重数据融合方法来估计标签脉冲响应的幅度和位置信息。4.成功地设计了一套完整的SAW射频识别系统,包括SAW标签和阅读器。并通过试验验证了本文提出的研究方法的正确性和可行性。更多还原

【Abstract】 Surface Acoustic Wave (SAW) tag is passive and can perform better in harsh environments than other passive RFID tag; so much attention has been paid to it.Theoretical analysis about SAW Tag system and reader is discussed. The emphasis of this paper was put on the improvement of identification probability, multi-tags identification and enhancement of accuracy in detection. A complete SAW tag system which includes SAW tag and reader is designed and manufactured. At first, theories on radiation field, antenna model, and antenna noise are introduced in this thesis.Secondly, different code methods of SAW tag are discussed and their merit and demerit are given. The influence of multi-reflection of SAW tag reflector is analyzed. The suggestion on selection of SAW tag material is proposed.Thirdly, the system structure of reader is discussed based on signal detection theory and free space propagation technique. The different composition and principle of time-domain sampling and frequency-domain sampling method are described and compared, as a conclusion, time-domain sampling method is chosen in reader designing here.Detection method of time-domain sampling is investigated, which comprises amplitude detection system, phase detection system and phase evaluation system. Amplitude detection can achieve higher detection probability when SNR is high, on contrary, frequency detection can achieve better detection probability when SNR is low. According with this fact, a fusion detection method depended on the working distance is proposed to improve the identification probability for single tag in this thesis.Fourthly, since SAW tags do not contain any intelligence, the question of multiple reads in the capture window comes up. The currently anti-collision methods are introduced. A new Walsh Matched-Threshold Filtering based anti-collision is proposed.Fifthly, the peak and position of reflected pulse are two main parameters of tag decoding. A weighted data fusion algorithm based on the matching pursuit (MP)-wavelet packet (WP) atomic decomposition method, which estimates the tag feature parameters of surface acoustic wave (SAW) is presented.Sixth, a total SAW tag system is designed successfully, which includes SAW tag and reader. The reader includes RF transceiver, analog-to-digital convertor with high speed, data processing etc. It can detect SAW tag signal by amplitude and phase. The improved radio channel can achieve more than 60dB isolation.Finally, test and experiment are made, the result indicates that the working distance reached over 4.5 meter by this own-made reader, it is proved also that the proposed theories and methods are correct and feasible.The creations performed in this thesis are:1. Fusion detection method depended on working distance is proposed to improve the identification probability for single tag;2. A new anti-collision method, the Walsh Matched-Threshold Filtering, used for multi-tags identification;3. To estimate the tag’s information presented by amplitude and position using weighted data fusion algorithm based on the matching pursuit(MP)-wavelet packet(WP) atomic decomposition method;4. Designing and manufacturing a complete tag reader system successfully.更多还原