基于克里金空间插值的位置指纹数据库建立算法
发布时间：2018-09-02 06:33
【摘要】：为解决室内定位系统中建立位置指纹数据库工作量庞大的问题,提出了一种融合信号衰减因素的普通克里金空间插值建库算法。该方法首先采用高斯滤波对有限预选参考点的信号强度采样数据进行预处理,并基于球状模型由参考点预处理数据拟合出空间变异函数;然后采用普通克里金插值法对其他位置的信号强度进行估值并生成相应的位置指纹;最后由有限实测数据生成大容量高分辨位置指纹数据库,并通过KNN_Filter算法和对数权重算法完成定位仿真,验证了该建库方法的有效性。仿真实验结果表明,该算法在保证定位精度的前提下,相比传统建库方法可降低40%左右的工作量,提高了室内位置指纹定位方法的工作效率。
[Abstract]:In order to solve the problem of the huge workload of establishing the location fingerprint database in the indoor positioning system, a common Kriging spatial interpolation database building algorithm is proposed, which combines the attenuation factors of the signal. Firstly, Gao Si filter is used to preprocess the signal intensity sampling data of the limited pre-selected reference points, and the spatial variogram is fitted from the pre-processed data of the reference points based on the spherical model. Then the common Kriging interpolation method is used to estimate the signal intensity of other positions and generate the corresponding location fingerprint. Finally, the large capacity high-resolution location fingerprint database is generated from the limited measured data. KNN_Filter algorithm and logarithmic weight algorithm are used to complete the localization simulation, and the validity of the method is verified. The simulation results show that the algorithm can reduce the workload by about 40% compared with the traditional database building method and improve the efficiency of the indoor location fingerprint location method on the premise of ensuring the location accuracy.
【作者单位】： 辽宁工业大学电子与信息工程学院;
【基金】：辽宁省科技厅博士启动基金资助项目(20131045) 辽宁省教育厅资助项目(L2012218)
【分类号】：TN911.7
本文编号：2218550
[Abstract]:In order to solve the problem of the huge workload of establishing the location fingerprint database in the indoor positioning system, a common Kriging spatial interpolation database building algorithm is proposed, which combines the attenuation factors of the signal. Firstly, Gao Si filter is used to preprocess the signal intensity sampling data of the limited pre-selected reference points, and the spatial variogram is fitted from the pre-processed data of the reference points based on the spherical model. Then the common Kriging interpolation method is used to estimate the signal intensity of other positions and generate the corresponding location fingerprint. Finally, the large capacity high-resolution location fingerprint database is generated from the limited measured data. KNN_Filter algorithm and logarithmic weight algorithm are used to complete the localization simulation, and the validity of the method is verified. The simulation results show that the algorithm can reduce the workload by about 40% compared with the traditional database building method and improve the efficiency of the indoor location fingerprint location method on the premise of ensuring the location accuracy.
【作者单位】： 辽宁工业大学电子与信息工程学院;
【基金】：辽宁省科技厅博士启动基金资助项目(20131045) 辽宁省教育厅资助项目(L2012218)
【分类号】：TN911.7
【相似文献】
相关期刊论文 前4条
1 曾浩然;张琳娜;赵凤霞;;新一代GPS区域滤波技术在表面结构计量中的应用研究[J];机械设计与制造;2011年04期
2 方依兰;高斯滤波最小移频键控调制方式[J];西南民族学院学报(自然科学版);1998年04期
3 郭斌;;高斯滤波调制解调器的实现技术[J];现代电子技术;2006年15期
4 ;[J];;年期
相关博士学位论文 前1条
1 李惠芬;基于新一代GPS体系的表面稳健高斯滤波技术的研究[D];华中科技大学;2004年
，本文编号：2218550
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2218550.html
