基于BP神经网络的智能轮胎标签仿真研究

轮胎中植入的RFID标签，可以长时间的很好的存储轮胎的型号、胎压、出厂日期等信息。RFID标签在空气中的阅读距离可以达到很大的距离，但是一旦植入轮胎中，很容易受到轮胎中的金属层和炭黑等电介质的影响，导致读取距离下降。所以，需要寻找合适的方法来预测不同RFID标签情况下的阅读器读取距离，就显得尤为重要了。 为了更加快捷方便的研究两者之间的关系，在天线长度、轮胎的介电常数、与钢丝层的距离都变化的情况下，利用FEKO电磁仿真软件建立了不同情况下的天线，并仿真得到反射系数S_11，然后利用弗林斯传输方程（Friis）计算得到仿真读取距离。MATLAB中有可供调用的神经网络工具箱，利用MATLAB强大的数据处理能力，建立BP神经网络预测模型，从而建立起标签天线长度、轮胎中标签与钢丝层的距离、轮胎介电常数和已得到的仿真读取距离之间的BP神经网络模型。实际测量值与训练后得到的预测仿真值在误差允许的范围内可以认定为实际测量距离。 因此，可以通过建立BP神经网络模型的方法，快速方便的在一定精度范围内预测阅读器的阅读距离。

Simulation to Intelligent Tire Tag Based on BP Neural Network

In order to study the relationship between RFID tag implantation environment and Readers read distance,thereby predicting the maximum reading distance of the RFID tag implanted into the tire;the FEKO electromagnetic simulation software established many antennas with different conditions and the reflection coefficient S_11 is obtained, And then use the Frings transfer equation (Friis) to calculate the simulation read distance. In MATLAB, there are neural network toolboxes that can be called, and the BP neural network prediction model is established by using MATLAB''s powerful data processing ability to establish the length of the tag antenna, the distance between the label and the wire layer in the tire, the tire dielectric constant and the simulation of the distance between the BP neural network model.;At last,The actual measured value and the predicted simulation value obtained after training can be regarded as the actual measurement distance within the allowable range of error;Therefore, the BP neural network model can be used to predict the reading distance of the reader within a certain precision range quickly and easily.