mmWave Beam-SNR Fingerprinting (mmBSF) for Precise Indoor Localization using Commercial-Off-The-Shelf (COTS) Routers

We describe our in-house dataset and an approach of fingerprinting-based indoor localization using COTS mmWave WiFi routers compliant with the IEEE802.11ad standards.

MERL Researchers: Toshiaki Koike-Akino, Pu (Perry) Wang, Milutin Pajovic, Philip V. Orlik. Joint work with MERL intern Haijian Sun (Utah State University, now Assistant Professor at University of Wisconsin-Whitewater).

Search MERL publications by keyword: mmWave, deep learning, localization, beamforming, fingerprinting,

Indoor Localization

This research aims to provide precise indoor localization without relaying of dedicated devices. The approach is one of WiFi sensing techniques, which make full use of ambient WiFi radio signals, considering the fact that our life is often surrounded by WiFi routers. Two WiFi channel attributes were considered by most fingerprinting-based indoor localization:

  • - coarse-grained received signal strength indicator (RSSI) measurements from the MAC layer
  • - fine-grained channel state information (CSI) from the physical layer

In this study, we introduce a new type of channel measurement in the mmWave band for fingerprinting-based indoor localization:

  • - mid-grained channel measurement: beam signal-to-noise ratios (SNRs)

These measurements are more informative (e.g., in the spatial domain) than RSSI measurements and easier to access than CSI measurements as the beam SNRs are required to be reported by users during mmWave beam training (therefore, no additional overhead is needed). Beam SNRs are defined in both the 5G and IEEE 802.11ad/ay standards.


In-house Measurement

To promote the use of mmWave beam SNR, we release the mmBSF open dataset collected using the MERL in-house testbed. We conducted a course of measurement sessions as follows:

  • - Three WiFi access points (APs): IEEE 802.11ad commercial routers
  • - Office environment at regular office hour
  • - Seven on-grid locations for fingerprinting at Day 1 (Locations 1, 2, ..., 7)
  • - Seven on-grid locations for testing at Day 2
  • - Four off-grid locations for testing at Day 3 (Locations A, B, C, D)
  • - Four orientations of handheld device (0, 90, 180, 270 degrees)


Deep Learning Localization

We demonstrated that deep neural network (DNN) based on residual network (ResNet) achieves accurate estimation of location area and user coordinate. Compared with conventional course-grained (RSSI-like single SNR) fingerprinting methods, Beam SNR fingerprinting offers prediction improvement greater than 2 fold in root-mean-square error.


Downloads

In-house mmWave beam-SNR fingerprinting dataset

  • - Training set (7 on-grid locations and 4 orientations): mmBSF_trainData.csv
  • - Testing set (7 on-grid locations and 4 orientations): mmBSF_testData_ongrid.csv
  • - Testing set (4 off-grid locations and 4 orientations): mmBSF_testData_offgrid.csv

Example script to load: mmBSF_LoS_dataLoad_visualization.ipynb

If you use or refer to this dataset, please cite the referenced papers.

Download dataset and script: mmBSF.zip



MERL Publications

  •  Koike-Akino, T., Wang, P., Pajovic, M., Sun, H., Orlik, P.V., "Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi: A Deep Learning Approach", IEEE Access, DOI: 10.1109/​ACCESS.2020.2991129, April 2020.
    BibTeX TR2020-054 PDF
    • @article{Koike-Akino2020apr,
    • author = {Koike-Akino, Toshiaki and Wang, Pu and Pajovic, Milutin and Sun, Haijian and Orlik, Philip V.},
    • title = {Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi: A Deep Learning Approach},
    • journal = {IEEE Access},
    • year = 2020,
    • month = apr,
    • doi = {10.1109/ACCESS.2020.2991129},
    • issn = {2169-3536},
    • url = {https://www.merl.com/publications/TR2020-054}
    • }
  •  Pajovic, M., Wang, P., Koike-Akino, T., Sun, H., Orlik, P.V., "Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi - Part I: RSS and Beam Indices", IEEE Global Communications Conference (GLOBECOM), DOI: 10.1109/​GLOBECOM38437.2019.9013466, December 2019.
    BibTeX TR2019-141 PDF
    • @inproceedings{Pajovic2019dec,
    • author = {Pajovic, Milutin and Wang, Pu and Koike-Akino, Toshiaki and Sun, Haijian and Orlik, Philip V.},
    • title = {Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi - Part I: RSS and Beam Indices},
    • booktitle = {IEEE Global Communications Conference (GLOBECOM)},
    • year = 2019,
    • month = dec,
    • publisher = {IEEE},
    • doi = {10.1109/GLOBECOM38437.2019.9013466},
    • issn = {2576-6813},
    • isbn = {978-1-7281-0962-6},
    • url = {https://www.merl.com/publications/TR2019-141}
    • }
  •  Wang, P., Pajovic, M., Koike-Akino, T., Sun, H., Orlik, P.V., "Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi - Part II: Spatial Beam SNRs", IEEE Global Communications Conference (GLOBECOM), DOI: 10.1109/​GLOBECOM38437.2019.9014103, December 2019.
    BibTeX TR2019-138 PDF
    • @inproceedings{Wang2019dec2,
    • author = {Wang, Pu and Pajovic, Milutin and Koike-Akino, Toshiaki and Sun, Haijian and Orlik, Philip V.},
    • title = {Fingerprinting-Based Indoor Localization with Commercial MMWave WiFi - Part II: Spatial Beam SNRs},
    • booktitle = {IEEE Global Communications Conference (GLOBECOM)},
    • year = 2019,
    • month = dec,
    • publisher = {IEEE},
    • doi = {10.1109/GLOBECOM38437.2019.9014103},
    • issn = {2576-6813},
    • isbn = {978-1-7281-0962-6},
    • url = {https://www.merl.com/publications/TR2019-138}
    • }