| 摘要: |
| 为减少室内定位复杂度并进一步提高定位精度，提出了一种5G超密集网络下的室内压缩重构指纹定位算法。该算法分为离线建库阶段和在线匹配阶段两个阶段。离线建库阶段采用了矩阵填充理论进行指纹库的构建，只需采取少量的指纹点构建具有低秩特性的局部指纹库，并通过非精确增广拉格朗日乘子法(Inexact Augmented Lagrangian Multiplier Method,IALM)算法进行矩阵填充，从而恢复完整的指纹库。在线匹配阶段采用卡方距离代替传统的欧式距离来计算待定位点与参考指纹点的相似度，并用加权K近邻算法估算出待定位点坐标。经过实验仿真分析，所提算法以1.13%的误差节约了40%的工作量，在信噪比为10 dB时定位误差最小为0.200 8 m,与传统K近邻指纹匹配算法相比具有更好的定位精度。 |
| 关键词: 5G超密集网络 室内定位 指纹定位 矩阵填充 加权K近邻算法 |
| DOI： |
|
| 基金项目:国家重点研发计划（2020YFC1511704）；国家自然科学基金资助项目（61971048）；北京市科技计划课题（Z191100001419012）；北京信息科技大学2020年促进高校内涵发展科研水平提高项目（2020KYNH212） |
|
| A compression and reconstruction fingerprint positioning algorithm for 5G indoor ultra-dense networks |
| JIA Pengfei,ZHANG Yuexia |
| (School of Information and Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China;a.School of Information and Communication Engineering;b.Key Laboratory of Modern Measurement & Control Technology,Ministry of Education;c.Beijing Key Laboratory of High Dynamic Navigation Technology,Beijing Information Science and Technology University,Beijing 100101,China) |
| Abstract: |
| In order to reduce the complexity of indoor positioning and further improve positioning accuracy,an indoor compression and reconstruction fingerprint positioning algorithm is proposed under 5G ultra-dense networks.The algorithm is divided into two stages,offline database building stage and online matching stage.In the offline database building stage,the matrix filling theory is used to construct the fingerprint database.Only a small number of fingerprint points are needed to construct a local fingerprint database with low-rank characteristics,and the matrix is filled by the Inexact Augmented Lagrangian Multiplier Method(IALM) algorithm to restore the complete fingerprint database.In the online matching stage,the Chi-square distance is used to replace the traditional Euclidean distance to calculate the similarity between the points to be located and the reference fingerprint points,and the Weighted K-nearest Neighbor(WKNN) algorithm is used to estimate the coordinates of the points to be located.According to the experimental simulation analysis,this algorithm saves 40% of the workload with an error of 1.13%,and the minimum positioning error is 0.200 8 m when signal-to-noise ratio is 10 dB,which has better positioning accuracy than the traditional K-nearest Neighbor(KNN) fingerprint matching algorithm. |
| Key words: 5G ultra-dense network indoor positioning fingerprint positioning matrix filling weighted K-nearest neighbor algorithm |
