DOI QR코드

DOI QR Code

A Deep Learning Based Device-free Indoor People Counting Using CSI

CSI를 활용한 딥러닝 기반의 실내 사람 수 추정 기법

  • An, Hyun-seong (Department of Electronic Engineering, Chungbuk National University) ;
  • Kim, Seungku (Department of Electronic Engineering, Chungbuk National University)
  • Received : 2020.04.01
  • Accepted : 2020.04.15
  • Published : 2020.07.31

Abstract

People estimation is important to provide IoT services. Most people counting technologies use camera or sensor data. However, the conventional technologies have the disadvantages of invasion of privacy and the need to install extra infrastructure. This paper proposes a method for estimating the number of people using a Wi-Fi AP. We use channel state information of Wi-Fi and analyze that using deep learning technology. It can be achieved by pre-installed Wi-Fi infrastructure that reduce cost for people estimation and privacy infringement. The proposed algorithm uses a k-binding data for pre-processing process and a 1D-CNN learning model. Two APs were installed to analyze the estimation results of six people. The result of the accurate number estimation was 64.8%, but the result of classifying the number of people into classes showed a high result of 84.5%. This algorithm is expected to be applicable to estimate the density of people in a small space.

사람 수 추정 기술은 IoT 서비스를 제공하기 위해 중요하다. 대부분의 사람 수 추정 기술은 카메라 또는 센서 데이터를 활용한다. 하지만 기존 기술들은 사생활 침해 문제가 발생 가능하며 추가로 인프라를 구축해야한다는 단점이 있다. 본 논문은 Wi-Fi AP를 활용하여 사람 수를 추정하는 방법을 제안한다. 사람 수 추정을 위해서 Wi-Fi의 채널 상태 정보를 딥러닝 기술을 활용하여 분석한다. Wi-Fi AP 기반 사람 수 추정 기술은 사생활 침해 우려가 없으며, 기존 Wi-Fi AP 인프라를 활용하면 되기 때문에 추가 비용이 발생하지 않는다. 제안하는 알고리즘은 k-바인딩 데이터 전처리 과정과 1D-CNN 학습 모델을 사용한다. AP 2대를 설치하여 6명의 사람 수 추정 결과를 실험을 통해 분석하였다. 정확한 사람 수 판별에 관한 결과는 64.8%로 낮은 결과를 보였지만, 사람의 수를 클래스로 분류한 결과는 84.5%의 높은 결과를 보였다. 해당 알고리즘은 제한된 공간에 사람의 밀집도를 파악하는데 응용 가능할 것으로 기대된다.

Keywords

References

  1. C. Kim, and S. Choi, "A Camera-Based System for Counting People in Real Time", Korea Institute Of Communication Sciences, vol. 2002, no. 66, pp. 503-506, 2002.
  2. S. Jang, and D. Jung, "Design of a People Counting System using Piezoelectric Sensors", The Korea Institute of Information and Communication Engineering, vol. 9, no. 1, pp. 149-152, 2017.
  3. H. Li, E. C. L. Chan, X. Guo, "Wi-Counter: SmartphoneBased People Counter Using Crowdsourced Wi-Fi Signal Data", IEEE Transactions on Human-Machine Systems, vol. 45, no. 4, pp. 442-452, 2015. https://doi.org/10.1109/THMS.2015.2401391
  4. T. Yoshida, and Y. Taniguchi, "Estimating the number of people using existing WiFi access point in indoor environment", in Proceedings of the 6th European Conference of Computer Science, Italy pp. 46-53, 2015.
  5. Y. Cheng, and R. Y. Chang, "Device-Free Indoor People Counting Using Wi-Fi Channel State Information for Internet of Things", in Proceedings IEEE Global Communications Conference, Singapore, pp. 1-6, 2017.
  6. S. D. Domenicon, M. D. Sanctis, and E. Cianca, "A trained-once crowd counting method using differential wifi channel state information", in Proceedings of the 3rd International on Workshop on Physical Analytics, Singapore, pp. 37-42, 2016.
  7. Atheros CSI Tool[Internet]. Available: https://wands.sg/research/wifi/AtherosCSI/
  8. TL-WDR4300[Internet]. Available: https://www.tp-link.com/kr/home-networking/wifi-router/tl-wdr4300/.
  9. keras Documentation[Internet]. Available: https://keras.io/.
  10. R. Crepaldi, J. Lee, R. Etkin, S. Lee, and R. Kravets, "CSI-SF: Estimating wireless channel state using CSI sampling & fusion" in Proceedings IEEE International Conference on Computer Communications. USA, pp. 154-162, 2012.