Establishment and optimization method of indoor/outdoor navigation context perception model based on GNSS and audio signals on the mobile phone

doi:10.13474/j.cnki.11-2246.2025.0218

Abstract

Abstract: The existing indoor/outdoor navigation context perception models based on machine learning and smart phones have the problems with only utilizing GNSS signals and poor perception accuracy. With the promotion of China's independent and controllable audio positioning technology, new available signals have been provided for that. In this paper, 12 GNSS and audio signal features on the mobile phone are selected, then an indoor/outdoor navigation context perception model based on whale optimization algorithm (WOA) and random forest (RF) is designed. The result shows that compared with only using audio signal features and GNSS signal features, the accuracy of context perception is significantly improved. Compared with the five traditional methods of back propagation neural network (BPNN), convolutional neural network (CNN), support vector machine (SVM), long short term memory (LSTM), and RF, the results of the proposed method are the best, accuracy, precision, recall rate and F1 are all exceeding 96%. The calculation speed of the proposed algorithm is basically equivalent to that of traditional RF.

Key words: indoor/outdoor navigation context detection, GNSS and audio signals, smart phone, random forest, whale optimization algorithm

CLC Number:

P228

LI Ang, ZHANG Xuedong, MA Yue, YANG Zhen, LI Linyang. Establishment and optimization method of indoor/outdoor navigation context perception model based on GNSS and audio signals on the mobile phone[J]. Bulletin of Surveying and Mapping, 2025, 0(2): 101-107.

References

[1] 陈惟杰,朱锋,郭斐,等.不同水深层的GNSS信号特征分析及其导航场景聚类划分[J].武汉大学学报(信息科学版),2024,49(1): 139-145.
[2] RADU V,KATSIKOULI P,SARKAR R,et al.A semi-supervised learning approach for robust indoor-outdoor detection with smartphones[C]//Proceedings of the 12th ACM Conference on Embedded Network Sensor Systems.Memphis Tennessee:ACM Press,2014:280-294.
[3] KIM D K,PARK J H,LEE J W.Sensing model for reducing power consumption for indoor/outdoor context transition[J].Journal of KIISE,2016,43(7):763-772.
[4] ZHOU P F,ZHENG Y Q,LI Z J,et al.IODetector: a generic service for indoor outdoor detection[J] ACM Transactions on Sensor Network,2024,11(2):36-362.
[5] 来奇峰,袁洪,魏东岩,等.基于场景检测的城市环境GNSS/INS组合定位方法研究[J].导航定位与授时,2021,8(1):151-162.
[6] 孟骞,姜颖颖,王立辉,等.城市峡谷环境多源导航信息弹性融合定位方法[J].导航与控制,2023,22(2):16-22.
[7] 杨元喜,杨诚,任夏.PNT智能服务[J].测绘学报,2021,50(8):1006-1012.
[8] HARIDASAN M,MOHOMED I,TERRY D,et al.StarTrack next generation: a scalable infrastructure for track-based applications[C]//Proceedings of the 9th USENIX Symposium on Operating Systems Design and Implementation,OSDI 2010.Vancouver,BC:[s.n.],2019:409-422.
[9] ZHU Y D,LUO H Y,WANG Q,et al.A fast indoor/outdoor transition detection algorithm based on machine learning[J].Sensors (Basel,Switzerland),2019,19(4):786.
[10] 周楠树.基于可穿戴设备的室内/室外场景识别研究[D].上海:上海交通大学,2017.
[11] GAO H,GROVES P D.Environmental context detection for adaptive navigation using GNSS measurements from a smartphone[J].Navigation,2018,65(1):99-116.
[12] OKAMOTO M,CHEN C.Improving GPS-based indoor-outdoor detection with moving direction information from smartphone[C]//Proceedings of 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2015 ACM International Symposium on Wearable Computers.Osaka:ACM Press,2015:257-260.
[13] LI S N,QIN Z,SONG H B,et al.A lightweight and aggregated system for indoor/outdoor detection using smart devices[J].Future Generation Computer Systems,2020,107: 988-997.
[14] 胡贤贤.基于智能手机的室内外场景识别系统研究与实现[D].徐州:中国矿业大学,2021.
[15] HÖFLINGER F,ZHANG R,HOPPE J,et al.Acoustic self-calibrating system for indoor smartphone tracking (ASSIST) [C]// Proceedings of 2012 International Conference on Indoor Positioning and Indoor Navigation (IPIN).Sydney,NSW:[s.n.],2012.
[16] 郭光毅.基于射频音频信号的智能手机多源混合定位方法研究[J].测绘学报,2022,51(3):474.