GNSS 제한 실외 환경에서 희소 RSS 샘플을 활용한 신경망 기반 위치 추정 기법

정성훈^*

해군 소령/해군사관학교 기계시스템공학과 부교수

Neural-network-based Localization in GNSS-denied Outdoor Environments with Sparse RSS Samples

Sunghoon Jung^*

LCDR, ROK Navy/Associate professor, Dept. of Mechanical System Engineering, Republic of Korea Naval Academy

Correspondence to: ^*Sunghoon Jung Dept. of Mechanical System Engineering, Republic of Korea Naval Academy 1 Jungwon-ro, Jinhae-gu, Changwon-si, Gyungsangnam-do, 51704, Republic of Korea Tel: +82-55-907-5320 E-mail: hun401@navy.ac.kr

초록

본 연구는 GNSS(global navigation satellite system) 신호가 제한되는 실외 전장 환경에서 적은 수의 RSS(received signal strength) 샘플만으로 위치를 추정하는 신경망 기반 기법을 제안한다. 제안 기법은 단일 경량 신경망 구조를 채택하여 라디오맵 구축이나 다수의 샘플 없이도 동작하며, 3GPP UMa 시뮬레이션에서 기존 방식 대비 RMSE가 30 % 이상 개선되었다. 특히 LOS 환경에서 최대 53.7 % 향상이 나타났으며, NLOS 환경에서도 꼬리 구간의 큰 오차를 억제하였다. 따라서 제안 기법은 제한된 시간·자원 조건에서도 신속하고 안정적인 위치 추정을 가능하게 한다.

Abstract

This study proposes a neural-network-based localization method for GNSS(global navigation satellite system)-denied outdoor battlefield environments using only a small number of RSS(received signal strength) samples. The proposed approach adopts a lightweight single-network structure that operates without radio-map construction or large sample sets. Simulation results under the 3GPP UMa scenario demonstrated more than 30 % improvement in RMSE compared with conventional methods. In particular, the method achieved up to 53.7 % gain in the LOS case and effectively suppressed large errors in the NLOS tail region. Therefore, the proposed method enables fast and reliable localization under limited time and resource conditions.

Keywords:

Localization, Received Signal Strength, Triangulation, GNSS-denied Environment, Neutal Network

키워드:

위치 추정, 수신 신호 세기, 삼변측량, GNSS 제한 환경, 신경망

Acknowledgments

본 연구는 해군사관학교 해양연구소 학술연구과제 연구비 지원으로 수행된 연구임.

References

M. A. Safar and H. S. Maghdid, “A Comprehensive Review of Indoor/Outdoor Localization Solutions in IoT Era: Research Challenges and Future Perspectives,” Computer Networks, Vol. 212, 109041, 2022. [https://doi.org/10.1016/j.comnet.2022.109041]
R. Liu, Q. Zhang, Y. Zhang, R. Zhang, and T. Meng, “Deep Learning-based Transmitter Localization in Sparse Wireless Sensor Networks,” Sensors, Vol. 24, No. 16, 5335, 2024. [https://doi.org/10.3390/s24165335]
Ç. Yapar, R. Levie, G. Kutyniok, and G. Caire, “Real-time Outdoor Localization Using Radio Maps: A Deep Learning Approach,” IEEE Transactions on Wireless Communications, Vol. 22, No. 12, pp. 9703–9717, 2023. [https://doi.org/10.1109/TWC.2023.3273202]
J. Gao, D. Wu, F. Yin, Q. Kong, L. Xu, and S. Cui, “MetaLoc: Learning to Learn Wireless Localization,” IEEE Journal on Selected Areas in Communications, Vol. 41, No. 12, pp. 3831–3847, 2023. [https://doi.org/10.1109/JSAC.2023.3322766]
Rappaport, T. S., Wireless Communications: Principles and Practice, 2nd ed. Upper Saddle River, NJ, USA: Prentice Hall, 2002.
3rd Generation Partnership Project (3GPP), Study on Channel Model for Frequencies from 0.5 to 100 GHz, 3GPP TR 38.901, V18.0.0, Release 18, Dec. 2023.