실내 핑거프린트 측위를 위한 딥러닝 기반의 초고해상도 RF 맵 재구성 기법 = A Deep Learning-Based Super-Resolution RF Map Reconstruction for Indoor Fingerprint Positioning

다국어 초록 (Multilingual Abstract)

In this paper, we propose a deep learning-based radio frequency (RF) map reconstruction method to improve the accuracy of the indoor fingerprint positioning. The proposed scheme reconstructs the RF map in super-resolution using a convolutional neural network (CNN) by directly learning with sparse data and ground truth data in the offline phase. The simulation results show that the proposed method has 7.53m improved positioning accuracy than the conventional fingerprint positioning and 0.92m additional positioning accuracy than the bicubic interpolation method.

국문 초록 (Abstract)

본 논문은 실내 핑거프린트 측위의 정확도를 향상시키기 위한 딥러닝 기반의 RF 맵 재구성 기법을 제안한다. 제안 기법은 합성곱 신경망 구조를 사용하여 희소 데이터와 정답 데이터와의 관계를 직접 학습함으로써 오프라인 단계에서 조사하지 않은 지역의 RF 맵을 고해상도로 복원한다. 실험 결과를 통해 제안 기법이 기존의 핑거프린트 측위보다 7.53m 향상된 측위 정확도를 가지며 다항식 보간법으로 RF 맵을 재구성하는 기법보다 0.92m의 추가적인 측위 정확도를 획득하였다.

참고문헌 (Reference)

1 S. He, "Wi-Fi fingerprint based indoor positioning : Recent advances and comparisons" 18 (18): 466-490, 2016

2 Simonyan, Karen, "Very deep convolutional networks for large-scale image recognition"

3 H. Liu, "Survey of wireless indoor positioning techniques and systems" 37 (37): 1067-1080, 2007

4 C. Wu, "Smartphones based crowdsourcing for indoor localization" 14 (14): 444-457, 2015

5 R. Bormann, "Room segmentation : Survey implementation and analysis" 1019-1026, 2016

6 Y. Ye, "RMapCS : Radio map construction from crowdsourced samples for indoor localiza tion" 6 : 24224-24238, 2018

7 X. Lin, "Positioning for the internet of things : A 3GPP perspective" 55 (55): 179-185, 2017

8 J. Koo, "Localizing WiFi access points using signal strength" 15 (15): 187-189, 2011

9 A. Makki, "Indoor localization using 802. 11 time differences of arrival" 65 (65): 614-623, 2016

10 Y. Zheng, "Exploiting AoA Estimation Accuracy for Indoor Localization : A Weighted AoA-Based Approach" 8 (8): 65-68, 2019

11 A. Borrelli, "Channel models for IEEE 802. 11b indoor system design" 3701-3705, 2004

12 K. Zhang, "Beyond a Gaussian denoiser : Residual learning of deep CNN for image denoising" 26 (26): 3142-3155, 2017

13 J. Kim, "Accurate image super-resolution using very deep convolutional networks" 1646-1654, 2016