摘要
为提高基于接收信号强度指示(RSSI)室内定位的定位精度,并降低时效性对定位的影响,提出将加权仿射传播聚类(WAP)与广义径向基函数(RBF)神经网络相结合的室内定位算法以及一种指纹数据优化方案。通过高斯分布对指纹数据进行优化;采用熵值法对仿射传播聚类算法的偏向参数p进行均值加权处理,得到粗定位结果;结合广义RBF神经网络得到修正后的定位结果。实验结果表明,该方法在提高室内定位精度的同时,减少了RSSI值的时效性对定位精度的影响。
To improve the positioning accuracy based on received signal strength indication(RSSI)indoor positioning and to reduce the effect of timeliness on positioning,an indoor location algorithm that combined weighted affine propagation clustering(WAP)and generalized radial basis function(RBF)neural networks was proposed,and a fingerprint data optimization scheme was also proposed.The fingerprint data were optimized by Gaussian distribution.The entropy method was used to average the bias parameter p of the affine propagation clustering algorithm to obtain the coarse positioning result.The generalized RBF neural network was combined to obtain the modified positioning result.Experimental results show that the proposed method improves the indoor positioning accuracy while reducing the effects of the timeliness of RSSI value on the positioning accuracy.
作者
宋宛真
冯秀芳
SONG Wan-zhen;FENG Xiu-fang(School of Information and Computer Science,Taiyuan University of Technology,Jinzhong 030600,China;School of Software,Taiyuan University of Technology,Jinzhong 030600,China)
出处
《计算机工程与设计》
北大核心
2021年第2期533-537,共5页
Computer Engineering and Design
基金
虚拟现实技术与系统国家重点实验室(北京航空航天大学)开放基金项目(VRLAB2019A05)。
关键词
RSSI
广义RBF神经网络
WAP算法
熵值法
指纹数据优化
RSSI
generalized RBF neural network
WAP algorithm
entropy value method
optimization of fingerprint database
