摘要
针对传统的无线信号路径损耗模型在预测距离值时易受环境参数A、n(A为1 m处的信号强度、n为路径损耗因子)影响的问题,提出遗传算法(GA)改进的反向传播(BP)神经网络构建无线信号路径损耗模型(GA-BP):分析基于BP神经网络构建无线信号路径损耗模型;然后利用遗传算法对BP神经网络中的初始权值和阈值进行优化,从而克服BP神经网络局部极小解的缺陷。实验结果表明,提出的GA-BP神经网络模型的测距精度比BP神经网络构建无线信号路径损耗模型的测距精度平均提高48%,并可避免对环境参数的依赖。
Aiming at the problem that it is liable to environmental parameters A and n(A represents the signal strength at 1 m and n represents the path loss factor)for the traditional wireless signal path loss model in the prediction of distance values,a wireless signal path loss model based on BP neural network optimized by genetic algorithm(GA)was proposed:the path loss model of wireless signal based on BP neural network was analyzed;and genetic algorithm was used to optimize the initial weights and thresholds of BP neural network,which helps overcome the defect of local minimum solution of BP neural network.Experimental result showed that the ranging accuracy of the proposed model could be average 48%higher than that of the wireless signal path loss model with BP neural network,meanwhile avoiding the dependence on environmental parameters.
作者
余振宝
卢小平
陶晓晓
周雨石
皇永波
YU Zhenbao;LU Xiaoping;TAO Xiaoxiao;ZHOU Yushi;HUANG Yongbo(Key Laboratory of Mine Spatial Information and Technology of NASMG,Henan Polytechnic University,Jiaozuo,Henan 454000,China)
出处
《导航定位学报》
CSCD
2020年第2期63-68,共6页
Journal of Navigation and Positioning
基金
国家重点研发计划项目(2016YFC0803103)
河南省高校创新团队支持计划项目(14IRTSTHN026)。
关键词
路径损耗模型
遗传算法
反向传播神经网络
测距精度
室内定位
path loss model
genetic algorithm(GA)
back propagation(BP)neural network
ranging accuracy
indoor location
