基于DBSCAN子空间匹配的蜂窝网室内指纹定位算法

发布时间：2018-03-02 14:48

本文选题：室内定位　切入点：蜂窝网　出处：《电子与信息学报》2017年05期 　论文类型：期刊论文

【摘要】：针对无线信道动态衰落特性引起的蜂窝网室内定位误差较大的问题,该文提出基于密度的空间聚类(Density Based Spatial Clustering of Applications with Noise,DBSCAN)子空间匹配算法,有效剔除大误差点,提高定位精度。首先通过划分信号空间,构建多个子空间,在子空间中利用加权K近邻匹配算法(Weighted K Nearest Neighbor,WKNN)估计出目标位置;然后利用DBSCAN对估计位置进行聚类以剔除异常点;最后结合概率模型确定最终估计位置。实验结果表明,基于DBSCAN的子空间匹配算法能有效剔除大误差点,提高蜂窝网室内定位系统的整体性能。
[Abstract]:In order to solve the problem of large indoor location error caused by the dynamic fading characteristics of wireless channels, this paper proposes a density-based spatial clustering Based Clustering of Applications with Noisegne DBSCAN-based subspace matching algorithm, which can effectively eliminate large error points. Firstly, several subspaces are constructed by dividing the signal space, and then the weighted K Nearest neighbor matching algorithm is used to estimate the target position in the subspace, and then the estimated position is clustered by DBSCAN to eliminate the outliers. The experimental results show that the subspace matching algorithm based on DBSCAN can effectively eliminate large error points and improve the overall performance of indoor positioning system in cellular networks.
【作者单位】：重庆邮电大学移动通信重点实验室;
【基金】：国家自然科学基金(61301126) 长江学者和创新团队发展计划(IRT1299) 重庆市基础与前沿研究计划(cstc2013jcyjA 40041,cstc2015jcyj BX0065) 重庆邮电大学青年科学研究项目(A2013-31)~~
【分类号】：TN929.53

【相似文献】