Indoor location algorithm based on improved AP selection and random forest fusion
-
摘要: 针对复杂室内环境下接收信号强度(RSS)值和维度发生变化的问题，提出一种改进的接入点(AP)选择方法并融合随机森林(RF)分类算法进行实时室内定位. 在离线阶段应用改进的AP选择方法，并使用AP的RSS数据方差以及AP出现频率来衡量AP稳定性并选取前m个稳定的AP. 在处理方差时会经拉普拉斯平滑，以避免出现方差为0的情况，并以此构建初步的指纹数据库；在在线阶段利用集成学习中的RF来对分类结果进行投票表决得到最终位置信息，并将改进后的算法同传统RF，改进后的AP选择融合加权的K近邻算法(WKNN)以及基于信息增益(IG)的AP选择算法加随机森林相比较. 实验结果表明：文中所提出的方法在定位误差方面较其他三个算法分别下降29.3%、23.2%、17.2%，同时在定位时间方面也有提升.
-
关键词:
- Wi-Fi室内定位 /
- 改进的接入点(AP)选择 /
- 随机森林(RF) /
- 位置指纹 /
- 拉普拉斯平滑
Abstract: Aiming at the problem of the received signal strength (RSS) value and dimension change in complex indoor environment, an improved access point (AP) selection method and a random forest (RF) classification algorithm for real-time indoor location are proposed. The improved AP selection method in the off-line phase uses the RSS data variance of the AP and the AP appearance frequency to measure the AP stability and selects the first m stable APs. When the variance is processed, the Laplacian smoothing is performed to avoid the variance of 0, and construct a preliminary fingerprint database. The online phase uses the RF in the integrated learning to vote on the classification result to arrive at the final position. The improved algorithm is compared with the traditional random forest, the improved AP selection fusion weighted K-nearest neighbor algorithm (WKNN) and the information gain (IG)-based AP selection algorithm plus random forest. The experimental results show that the proposed method Compared with the other three algorithms, the positioning error decreased by 29.3%, 23.2%, and 17.2%, respectively, and the positioning time is also improved. -
表 1 实验采集的部分RSS数据
dBm 定位区域 部分定位指纹数据 坐标1处的RSS数据 −90，−88，−92，−96，−89，−90，−92，−85，−96，
−97，−99，−100，−97，−99，−92，−89，−80，−87，
−80，−84，−72，−77，−76，−83，−77，−73，−70，
−56，−65，−79，−66，−50，−49，−66，−59，−60，
−52，−57，−59，−54，−54，−60，−80，−85，−85，
−94，−74，−85坐标2处的RSS数据 −72，−70，−69，−75，−74，−80，−67，−73，−83，
−80，−81，−81，−85，−87，−79，−94，−93，−80，
−90，−88，−73，−89，−84，−84，−92，−95，−67，
−60，−72，−68，−64，−88，−82，−76，−89，−84，
−76，−75，−89，−82，−84，−75，−73，−71，−79，
−70，−78，−69，−78，−75
下载: 导出CSV
-
[1] GU Y Y, LO A, NIEMEGEERS I. A survey of indoor positioning systems for wireless personal networks[J]. IEEE communications surveys and tutorials, 2009, 11(1): 13-32. DOI: 10.1109/SURV.2009.090103 [2] BAHL P, PADMANABHAN V N. RADAR: an in-building RF-based user location and tracking system[C]// The 19th Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064), 2020. DOI: 10.1109/INFCOM.2000.832252 [3] 张鹤丹. 基于WiFi技术的井下人员定位系统研究[D]. 西安: 西安建筑科技大学, 2013. [4] 李新春, 侯跃. 基于改进AP选择和K最近邻法算法的室内定位技术[J]. 计算机应用, 2017, 37(11): 3276-3280, 3287. [5] 罗宇锋, 候利军. 多方位分组式 WiFi 室内定位算法[J]. 传感器与微系统, 2021, 40(4): 139-141. [6] 郭妍, 陈晓, 任晓晔. 一种优化随机森林模型的室内定位方法[J]. 激光杂志, 2018, 39(10): 70-74. [7] 韩学法, 吴飞, 朱海, 等. 基于 GF-KF修正RSSI的室内指纹定位方法[J]. 全球定位系统, 2020, 45(3): 54-62. [8] LEE S M, KIM J, MOON N. Random forest and Wi-Fi fingerprint-based indoor location recognition system using smart watch[J]. Human-centric computing and information sciences, 2019, 9(1): 6. DOI: 10.1186/s13673-019-0168-7 [9] 沈阳. 基于指纹的无线室内定位中接入点选择算法研究[D]. 杭州: 浙江大学, 2014. [10] TANIUCHI D, MAEKAWA T. Robust Wi-Fi based indoor positioning with ensemble learning[C]//IEEE the 10th International Conference on Wireless and Mobile Computing, Networking and Communications, 2014. DOI: 10.1109/WiMOB.2014.6962230 [11] 张伟, 花向红, 邱卫宁, 等. WiFi指纹定位的一种新组合算法[J]. 测绘工程, 2017, 26(3): 14-18. [12] 刘少伟, 花向红, 邱卫宁, 等. WiFi指纹定位中AP个数对定位精度的影响[J]. 测绘工程, 2017, 26(2): 33-36. [13] 陈伟纲. 采用非自主部署AP的WLAN室内定位指纹库关键技术研究[D]. 广州: 华南理工大学, 2016. [14] BREIMAN L. Random forests[J]. Machine learning, 2001, 45(1): 5-32. DOI: 10.1023/A:1010933404324 -
点击查看大图
图(7) / 表(1)
计量
- 文章访问数: 425
- HTML全文浏览量: 192
- PDF下载量: 23
- 被引次数: 0
