RSSI室内定位在线匹配算法的研究与性能比较
The Research and Performance Comparison of RSSI Indoor Positioning Online Matching Algorithms
吴之宁 1汪学刚 1邹林1
作者信息
- 1. 电子科技大学信息与通信工程学院,四川 成都 611730
- 折叠
摘要
针对在基于WiFi信号强度RSSI进行室内定位的指纹库算法的在线匹配环节中存在的不足,该文利用基于阈值R0 动态筛选匹配的指纹点数,提出了一种增强加权k近邻算法(EWKNN).因为阈值R0 可以动态筛选指纹库中的样本点,所以能够提高增强加权k近邻算法的适用度和高精度.仿真结果表明:在R0 设置恰当的情况下,增强加权k近邻算法的计算量与加权k近邻算法(WKNN)相当,但定位精度更高.
Abstract
Focused on the online matching part in fingerprint database algorithm for indoor positioning based on WiFi signal strength RSSI,the enhanced weight k-nearest method is proposed by dynamically selecting the matching fingerprint points based on the threshold R0.The effectiveness of the enhanced weighted k-nearest neighbors algo-rithm(EWKNN)stems from the threshold,because the value of R0 can dynamically filter the sample points in the fingerprint library,which is an improvement on the weight k-nearest neighbors algorithm.The result of the simulation shows that under the appropriate setting of R0,the amount of calculation of the enhanced weight k-nearest neighbors algorithm(EWKNN)is comparable to the weighted k-nearest neighbor algorithm(WKNN),but the positioning accu-racy is higher.
关键词
室内定位/指纹库在线匹配/增强加权k近邻算法/加权k近邻算法/累积分布函数Key words
indoor positioning/the online matching of fingerprint database algorithm/enhanced weight k-nearest neighborhood algorithm/weight k-nearest neighborhood algorithm/cumulative distribution function引用本文复制引用
基金项目
国家自然科学基金重大仪器专项(42027805)
出版年
2024
NETL
NSTL
万方数据