Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 May 11:10:1106439.
doi: 10.3389/frobt.2023.1106439. eCollection 2023.

Using convolutional neural networks to detect GNSS multipath

Anthony Guillard  1   2 , Paul Thevenon  1 , Carl Milner  1
Affiliations

Using convolutional neural networks to detect GNSS multipath

Anthony Guillard et al. Front Robot AI. .

Abstract

Global Navigation Satellite System (GNSS) multipath has always been extensively researched as it is one of the hardest error sources to predict and model. External sensors are often used to remove or detect it, which transforms the process into a cumbersome data set-up. Thus, we decided to only use GNSS correlator outputs to detect a large-amplitude multipath, on Galileo E1-B and GPS L1 C/A, using a convolutional neural network (CNN). This network was trained using 101 correlator outputs being used as a theoretical classifier. To take advantage of the strengths of convolutional neural networks for image detection, images representing the correlator output values as a function of delay and time were generated. The presented model has an F score of 94.7% on Galileo E1-B and 91.6% on GPS L1 C/A. To reduce the computational load, the number of correlator outputs and correlator sampling frequency was then decreased by a factor of 4, and the convolutional neural network still has an F score of 91.8% on Galileo E1-B and 90.5% on GPS L1 C/A.

Keywords: DLL; Global Navigation Satellite System; convolutional neural network; correlator; machine learning; multipath.

PubMed Disclaimer

Conflict of interest statement

Author AG was employed by the Company 3D Aerospace. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Early–late power discriminator function in error-free conditions (left) and affected by some multipath (right).
FIGURE 2
FIGURE 2
Multipath error envelope for a brick-induced multipath for an EL spacing of 0.04 chips on Galileo E1-B (left) and for an EL spacing of 0.125 chips on GPS L1 C/A (right).
FIGURE 3
FIGURE 3
PRN deformation correction algorithm.
FIGURE 4
FIGURE 4
CNN architecture for multipath detection.
FIGURE 5
FIGURE 5
Illustration of some correlator variations for GPS L1 C/A.
FIGURE 6
FIGURE 6
Filtered (black) and non-filtered (blue) tracking bias with its multipath bounds (red) for Galileo E1-B on PRN 3 (left) and their corresponding CNN inputs (right).
FIGURE 7
FIGURE 7
Accuracy and loss values over training iterations for training and validation datasets.
FIGURE 8
FIGURE 8
E1 metrics of the CNN.
FIGURE 9
FIGURE 9
L1 metrics of the CNN.
See this image and copyright information in PMC

References

    1. Abadi M., Barham P., Chen J., Chen Z., Davis A., Dean J., et al. (2016). “Tensorflow: A system for large-scale machine learning,” in 12th USENIX symposium on operating systems design and implementation (OSDI 16; ), Savannah Georgia, USA, 265–283.
    1. Agarap A. F. (2018). Deep learning using rectified linear units (relu). CoRR, abs/1803.08375.
    1. Bellad V., Petovello M. (2013). “Indoor multipath characterization and separation using distortions in gps receiver correlation peaks,” in ION GNSS+ 2013, Nashville, Tennessee, September 16-20, 2013.
    1. Bétaille D., Peyret F., Ortiz M., Miquel S., Fontenay L. (2013). A new modeling based on urban trenches to improve GNSS positioning quality of service in cities. Intell. Transp. Syst. Mag. IEEE 5, 59–70. 10.1109/mits.2013.2263460 - DOI
    1. Blais A., Couellan N., Munin E. (2022). A novel image representation of gnss correlation for deep learning multipath detection. Array 14, 100167. 10.1016/j.array.2022.100167 - DOI

LinkOut - more resources