摘要
提出了一种基于VGG-19网络和卡尔曼预处理的无线传感器网络(WSNs)测距方法。首先,通过卡尔曼滤波方法实现接收信号强度指示(RSSI)数据的滤波处理过程,以提高方法的稳定性;然后,利用VGG-19网络学习分析真实场景下拍摄的图像信息,以获得高精度的图像特征;最后,根据图像数据和实际信号衰减数据,自动提取不同地面环境下信号传播的高层语义特征,从而完成距离测量。基于实测平台获取的数据进行实验分析,结果表明:在4种复杂程度不一的场景下,所提模型的距离估计误差最大不超过1 m,并且测距耗时为34.92 ms,整体性能优于其他对比模型。
A wireless sensor networks(WSNs)ranging method based on VGG-19 network and Kalman preprocessing is proposed.Firstly,Kalman filtering method is used for filtering processing of received signal strength indication(RSSI)data,so as to improve the stability of the method.Then,VGG-19 network learning is used to analyze on image information captured in real scenes,so as to obtain high-precision image features.Finally,according to the image data and the actual signal attenuation data,high-level semantic features of signal propagation in different ground environments are automatically extracted,so as to complete distance measurement.Experimental analysis on data acquired by test platform is carried out.The experimental results show that the maximum distance estimation error of the proposed model is less than 1 m,and the ranging time is 34.92 ms,overall performance is better than other comparison models in four different complicated degrees sites.
作者
刘超敏
胡玉平
LIU Chaomin;HU Yuping(School of Information Engineering,Zhengzhou College of Finance and Economics,Zhengzhou 450000,China;School of Information Science,Guangdong University of Finance and Economics,Guangzhou 510320,China)
出处
《传感器与微系统》
CSCD
北大核心
2023年第10期139-142,共4页
Transducer and Microsystem Technologies
基金
广东省自然科学基金资助项目(2016A030313717)。
关键词
无线传感器网络
卡尔曼滤波处理
VGG-19网络
接收信号强度指示
高层语义特征
测距
wireless sensor network(WSNs)
Kalman filtering processing
VGG-19 network
received signal strength indication(RSSI)
high level semantic features
ranging
