Trajectory Estimation of Mobile Users/Devices based on Wi-Fi Fingerprinting and Deep Neural Networks

Table of Contents

Abstract

Where there is no GPS signal, received signal strength (RSS) from wireless network can be used for location estimation through fingerprinting; for instance, a vector of a pair of a service set identifier (SSID) and RSS for a Wi-Fi access point (AP) measured at a known location becomes its location fingerprint and a static user/device location then can be estimated by finding the closest match between its new RSS measurement and the location fingerprints in a database. This project aims at extending Wi-Fi fingerprinting technique to trajectory estimation of mobile users/devices exploiting its space/time correlations using deep neural networks (DNNs).

Research questions

We are to extend the Wi-Fi fingerprinting technique for the trajectory estimation of mobile users/devices using DNNs.

Below are key research questions in this regard:

  • How to exploit the spatial and temporal correlations of a sequence of locations of a mobile user in estimating its trajectory?
  • What are the impacts of the depth of dependencies (i.e., the number of past states that the current state depends on) in modelling a trajectory?
  • Which one is better in modelling the spatial and temporal correlations of a trajectory, Markov models or recurrent neural networks (RNNs)?

scalable_dnn_classifier.png

Figure 1: A DNN architectures for scalable building/floor classification and floor-level coordinates estimation based on an SAE for the reduction of feature space dimension and a feed-forward classifier for multi-label classification [8].

trajectory_estimation.png

Figure 2: Illustration of trajectory estimation based on the location estimates from Wi-Fi finger printing.

People

Research Assistant

Participants

Grants

Duration

  • Jun./2018–Aug./2018 (10 weeks)

Meetings

Outcomes

Publications

  • Zhenghang Zhong, Zhe Tang, Xiangxing Li, Tiancheng Yuan, Yang Yang, Wei Meng, Yuanyuan Zhang, Renzhi Sheng, Naomi Grant, Chongfeng Ling, Xintao Huan, Kyeong Soo Kim and Sanghyuk Lee, "XJTLUIndoorLoc: A new fingerprinting database for indoor localization and trajectory estimation based on Wi-Fi RSS and geomagnetic field," accepted for presentation at GCA'18, Sep. 27, 2018.

References

Trajectory estimation

  1. Ho Jun Jang, Jae Min Shin, and Lynn Choi, "Geomagnetic field based indoor localization using recurrent neural networks," Proc. GLOBECOM 2017, pp. 1-6, Dec. 2017. (DOI)
  2. Wei Zhang, Kan Liu, Weidong Zhang, Youmei Zhang, and Jason Gu, "Deep neural networks for wireless localization in indoor and outdoor environments," Neurocomputing, vol. 194, pp. 279-287, 2016. (DOI)

Recurrent neural networks (RNNs) and long short-term memory (LSTM) networks

  1. Christopher Olah, Understanding LSTM networks, Accessed May 22, 2018.
  2. Yuan Lukito and Antonius Rachmat Chrismanto, "Recurrent neural networks model for WiFi-based indoor positioning system," Proc. of 2017 International Conference on Smart Cities, Automation & Intelligent Computing Systems, Yogyakarta, Indonesia, Nov. 2017. (DOI)

Convolutional neural networks (CNNs) for time series data (e.g., audio signal)

  1. Alex Graves, Abdel-rahman Mohamed, and Geoffrey Hinton, "Speech recognition with deep recurrent neural networks," arXiv:1303.5778 [cs.NE], Mar. 2013. (arXiv)
  2. Grégoire Montavon, "Deep learning for spoken language identification," Proc. NIPS Workshop on Deep Learning for Speech Recognition and Related Applications, 2009.

Wi-Fi fingerprinting and deep learning

  1. Kyeong Soo Kim, Ruihao Wang, Zhenghang Zhong, Zikun Tan, Haowei Song, Jaehoon Cha, and Sanghyuk Lee, "Large-scale location-aware services in access: Hierarchical building/floor classification and location estimation using Wi-Fi fingerprinting based on deep neural networks," (Extended version of the FOAN 2017 paper), Fiber and Integrated Optics, vol. 37, no. 5, pp. 277-289, Apr. 27, 2018. (DOI)
  2. Kyeong Soo Kim, Sanghyuk Lee, and Kaizhu Huang "A scalable deep neural network architecture for multi-building and multi-floor indoor localization based on Wi-Fi fingerprinting," Big Data Analytics, vol. 3, no. 4., pp. 1017, Apr. 19, 2018. (DOI) (arXiv)
  3. Kyeong Soo Kim, Ruihao Wang, Zhenghang Zhong, Zikun Tan, Haowei Song, Jaehoon Cha, and Sanghyuk Lee, "Large-scale location-aware services in access: Hierarchical building/floor classification and location estimation using Wi-Fi fingerprinting based on deep neural networks," Proc. of FOAN 2017, Munich, Germany, Nov. 7, 2017. (arXiv)
  4. Mai Ibrahim, Marwan Torki, and Mustafa ElNainay, "CNN based indoor localization using RSS time-series," Proc. of ISCC 2018, pp. 1-6, Natal, Brazil, Jun. 2018. (ResearchGate)
  5. Paramvir Bahl and Venkata N. Padmanabhan, "RADAR: An In-Building RF-based User Location and Tracking System," Proc. INFOCOM 2000, pp. 1-10, 2000.
  6. Michal Nowicki and Jan Wietrzykowski, "Low-effort place recognition with WiFi fingerprints using deep learning," arXiv:1611.02049 [cs.RO], Apr. 2017. (arXiv)
  7. Keras: A high-level neural networks API, written in Python and capable of running on top of either TensorFlow or Theano; now part of TensorFlow distribution.
  8. Shixiong Xia, Yi Liu, Guan Yuan, Mingjun Zhu, and Zhaohui Wang, "Indoor Fingerprint Positioning Based on Wi-Fi: An Overview," ISPRS Int. J. Geo-Inf, vol. 6, no. 5, 135, pp. 1-25, 2017.
  9. John S. Breese, David Heckerman, and Carl Kadie, "Empirical analysis of predictive algorithms for collaborative filtering," Tech. Report, MSR-TR-98-12, Microsoft, May 1, 1998.
  10. CS246 Mining Massive Data Sets Winter 2017, Chapter 9. Recommendation systems, Stanford, 2017.

Fingerprint datasets

  1. J. Torres-Sospedra et al., "UJIIndoorLoc: A new multi-building and multi-floor database for WLAN fingerprint-based indoor localization problems," Proc. IPIN 2014, pp. 261-270, Busan, Korea, Oct. 2014. (DOI)
  2. J. Torres-Sospedra et al., "UJIIndoorLoc-Mag: A new database for magnetic field-based localization problems," Proc. IPIN 2015, pp. 1-10, Banff, Alberta, Canada, Oct. 2015. (DOI)
  3. UCI Machine Learning Repository, UJIIndoorLoc Data Set (fingerprint database for the above two papers and many others).
  4. E. S. Lohan et al., "Wi-Fi crowdsourced fingerprinting dataset for indoor positioning," Data, vol. 2, no. 4, article no. 32, pp. 1-16, 2017.

GitHub repositories for Python codes and fingerprint data

Android programming

Related Projects

Date: Time-stamp: <2022-07-15 Fri 21:52> Time-stamp: <2022-08-24 Wed 21:59>

Created: 2024-04-11 Thu 15:08

Emacs 29.1 (Org mode 9.6.6)

Validate