Effectiveness of Neural Network Approaches for the Acquisition of Non-Periodic Spreading Codes

Marco Trombini, Davide Leone, Angelo Bruno, Marco D’Addezio, Gianluca Falco, Emanuela Falletti

Abstract: The first operation performed by a GNSS receiver is signal acquisition. The goal is to assess whether the signal from a particular satellite is present or not, along with providing a first estimate of code delay and Doppler frequency. The solutions that are currently deployed in modern receivers are well-engineered and extensively field-tested. However, as nominal conditions not always hold in real applications, such classical techniques may poorly perform. This paper introduces the use of a deep learning-based approach to the signal acquisition problem, acknowledging the adaptability of such algorithms to the characteristics of the available data. The proposed pipeline leverages on Long-Short Term Memory Recurrent Neural Network, to detect satellite signals from the Cross Ambiguity Function. The experimental phase is conducted on simulated aperiodic GNSS signals referring to Galileo E1 samples, to highlight the real-world applicability of the proposed approach. The present method is compared with the classical approach from the literature, aimed at paving the way for the use of deep-learning based receivers in actual scenarios.
Published in: Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023)
September 11 - 15, 2023
Hyatt Regency Denver
Denver, Colorado
Pages: 206 - 215
Cite this article: Trombini, Marco, Leone, Davide, Bruno, Angelo, D’Addezio, Marco, Falco, Gianluca, Falletti, Emanuela, "Effectiveness of Neural Network Approaches for the Acquisition of Non-Periodic Spreading Codes," Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023), Denver, Colorado, September 2023, pp. 206-215. https://doi.org/10.33012/2023.19312
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In