基于改进克里金插值的位置指纹定位
首发时间:2018-04-10
摘要:为解决位置指纹定位系统中指纹采集工作量大的问题,提出一种基于共栖生物搜索克里金插值的位置指纹定位算法。首先在定位区域内稀疏选定参考点,采用高斯滤波对参考点上接收的信号强度进行异常值过滤;然后基于参考点采集数据结合克里金插值算法对非参考点的信号强度进行估计,其中利用共栖生物搜索算法优化球状模型参数,提高插值算法中理论变异函数的拟合精度;最后,由有限参考点实测数据生成密集且高分辨的位置指纹库,并结合WKNN算法完成定位仿真,验证该建库方法的有效性。实验结果表明,在保证定位精度的前提下,该建库算法较传统建库方法减小了约50%的采集工作量。
For information in English, please click here
Fingerprint Location Algorithm Based on Improved Kriging Interpolation
Abstract:In order to solve a problem about the huge workload of fingerprint sampling in the positioning system, this paper proposed a novel algorithm, which based on the Kriging interpolation of symbiotic organisms search. Firstly, this method performed sparse sampling on the positioning area and selected the sampling nodes as the interpolation reference nodes, then filtered out the signal strength outliers by the Gaussian filter. Secondly, it used the Kriging interpolation algorithm to estimate the signal strength of non-reference nodes based on the reference node data, where it used the symbiotic organisms search to optimize the parameters of the spherical model, improving the fitting precision of the theoretical variogram in the interpolation algorithm. Finally,it generated a dense and high-resolution location fingerprint database from the limited reference node data, and completed the positioning simulation by the WKNN algorithm to prove the validity of the database construction method. The experimental results show that compared with the traditional database establishment method, the proposed algorithm can reduce the sampling workload by about 50% under the premise of ensuring the positioning accuracy.
Keywords: indoor fingerprint positioning kriging interpolation symbiotic organisms search
基金:
引用
No.****
同行评议
共计0人参与
勘误表
基于改进克里金插值的位置指纹定位
评论
全部评论0/1000