Assessment of indoor positioning quality based on Bluetooth Low Energy beacons and machine learning

The paper addresses indoor positioning using Bluetooth Low Energy (BLE) beacons in environments where satellite navigation signals are unavailable. The study evaluates the positioning accuracy that can be achieved by combining BLE beacons with machine learning methods. A feedforward neural network with two hidden layers is used to estimate the 2D coordinates of an object. Three approaches to generating the training dataset are considered: a regular grid, a random grid, and a line-based grid. Signal propagation is modeled using a logarithmic path loss model with measurement noise. Experimental studies are conducted for indoor environments of different sizes and with different numbers of beacons. The results demonstrate that the proposed approach achieves a positioning accuracy of 20–30 cm, with the best performance obtained using regular or line-based training grids. It is shown that machine learning does not improve positioning accuracy in a statistical sense, and the problem remains within the framework of the fundamental concepts of observability and stability. The proposed method can be applied to indoor navigation of service robots, drones, automated warehouse systems, and safety monitoring applications.

1. Shchekotov M., Kashevnik A. Comparative analysis of indoor positioning systems for smartphones. SPIIRAS Proceedings. 2012;(4):459–471. (In Russ.). https://doi.org/10.15622/sp.23.26

2. Namiot D. On indoor positioning. International Journal of Open Information Technologies. 2015;3(3):23–26.

3. Davidson P., Kirkko-Jaakkola M., Collin J., Takala J. Navigation algorithm combining building plans with autonomous sensor data. Giroskopiya i navigatsiya. 2015;(1):29–42. (In Russ.).

4. Bolotova S.Yu., Zonov A.V., Tutin A.P. Indoor navigation in mobile applications. Software Engineering. 2018;9(1):29–34. (In Russ.). https://doi.org/10.17587/prin.9.29-34

5. Shchekotov M.S. SLAM method of indoor navigation based on bluetooth beacon localization. Journal of Information Technologies and Computing Systems. 2021;(3):70–81. (In Russ.). https://doi.org/10.14357/20718632210307

6. Evtushenko E.A., Kovalets I.A., Safaryan O.A. Development of an indoor positioning system. Trudy Severo‑Kavkazskogo filiala Moskovskogo tekhnicheskogo universiteta sviazi i informatiki. 2025;(1):44–45. (In Russ.).

7. Huang J., Junginger S., Liu H., Thurow K. Indoor Positioning Systems of Mobile Robots: A Review. Robotics. 2023;12(2):47. https://doi.org/10.3390/robotics12020047

8. Cao T. Generation of an indoor 2D map and track encryption based on mobile crowdsourcing. International Journal of Engineering and Technology. 2018;7(3):4–19. https://doi.org/10.14419/ijet.v7i3.19.16982

9. Kronenwett N., Ruppelt J., Trommer G.F. Motion monitoring based on a finite state machine for precise indoor localization. Giroskopiya i navigatsiya. 2017;25(1):33–48. (In Russ.).

10. Tomažič S. Indoor Positioning and Navigation. Sensors. 2021;21(14):4793. https://doi.org/10.3390/s21144793

11. Anbarasu B., Anitha G. Indoor Scene recognition for Micro Aerial Vehicles Navigation using Enhanced SIFT-ScSPM Descriptors. Journal of Navigation. 2020;73(1):37–55. https://doi.org/10.1017/S0373463319000420

12. Wu F., Liang Y., Fu Y., Geng C. A New Indoor Positioning System Using Artificial Encoded Magnetic Fields. Journal of Navigation. 2018;71(2):299–316. https://doi.org/10.1017/S0373463317000698

13. Li C., Huang H., Liao B. An Improved Fingerprint Algorithm with Access Point Selection and Reference Point Selection Strategies for Indoor Positioning. Journal of Navigation. 2020;73(6):1182–1201. https://doi.org/10.1017/S0373463319000730

14. Zhao J., Gao X., Wang X., Li C., Song M., Sun Q. An Efficient Radio Map Updating Algorithm based on K-Means and Gaussian Process Regression. Journal of Navigation. 2018;71(5):1055–1068. https://doi.org/10.1017/S037346331800019X

15. Kokoreva E.V. Application of neural network to position a mobile object indoors. Ekonomika i kachestvo sistem sviazi. 2025;(1):84–92. (In Russ.).

16. Lizneva Y.S., Kostyukovich A.E., Kokoreva E.V. Analysis of the possibilities of determining location in a Wi-Fi network using neural network algorithms. Software Systems and Computational Methods. 2024;(4):1–12. https://doi.org/10.7256/2454-0714.2024.4.72107

17. Grinyak V.M. 3D indoor positioning system based on Bluetooth beacons. Modeling, Optimization and Information Technology. 2020;8(3). (In Russ.). https://doi.org/10.26102/2310-6018/2020.30.3.023

18. Grinyak V.M., Devyatisilny A.S., Shurygin A.V. Feasibility Study of Indoor Navigation with Bluetooth Beacons. Information Technologies. 2018;24(9):610–617. (In Russ.). https://doi.org/10.17587/it.24.610-617

19. Meditch J.S. Stochastic optimal linear estimation and control. Moscow: Energiya; 1973. 440 p. (In Russ.).

20. Stepanov O.A. Fundamentals of estimation theory with applications to problems of navigation information processing. Part 1. An introduction to the theory of evaluation. Saint Petersburg: Central Scientific Research Institute ElectroPribor; 2017. 509 p. (In Russ.).

21. Stepanov O.A. Fundamentals of estimation theory with applications to problems of navigation information processing. Part 2. An introduction to the theory of filtration. Saint Petersburg: Central Scientific Research Institute ElectroPribor; 2017. 428 p. (In Russ.).