基于消费者WiFi的材料缺陷检测

Material Defect Detection Based on Consumer WiFi

本文提出了一种基于WiFi信号的材料缺陷检测系统,通过使用廉价的消费者WiFi设备来检测材料是否存在缺陷以及是否符合标准。材料缺陷检测系统通过信道状态信息来感知环境变化。通过从信道状态信息流中提取出有用信息,并使用主成分分析法(PCA)及巴特沃斯滤波器进行降噪,并进行特征提取,最终使用SVM算法对数据进行分类。在实验场景中,选用了建材作为实验材料,通过仿真发现,缺陷检测的真阳性率达到了92.22%。

This paper proposes a material defect detection system based on WiFi signals,which uses inexpensive consumer WiFi devices to detect whether materials have defects and meet standards.Material defect detection systems perceive environmental changes through channel state information.By extracting useful information from the channel state information stream,using principal component analysis(PCA)and Butterworth filter for noise reduction,and performing feature extraction,finally using SVM algorithm to classify the data.In the experimental scenario,building materials were selected as experimental materials.Through simulation,it was found that the true positive rate of defect detection reached 92.22%.