NLOS Multipath Classification of GNSS Signal Correlation Output Using Machine Learning

Taro Suzuki; Yoshiharu Amano

doi:10.3390/s21072503

NLOS Multipath Classification of GNSS Signal Correlation Output Using Machine Learning

Sensors (Basel). 2021 Apr 3;21(7):2503. doi: 10.3390/s21072503.

Authors

Taro Suzuki¹, Yoshiharu Amano²

Affiliations

¹ Future Robotics Technology Center, Chiba Institute of Technology, Chiba 2750016, Japan.
² Department of Applied Mechanics and Aerospace Engineering, Waseda University, Tokyo 1620044, Japan.

Abstract

This paper proposes a method for detecting non-line-of-sight (NLOS) multipath, which causes large positioning errors in a global navigation satellite system (GNSS). We use GNSS signal correlation output, which is the most primitive GNSS signal processing output, to detect NLOS multipath based on machine learning. The shape of the multi-correlator outputs is distorted due to the NLOS multipath. The features of the shape of the multi-correlator are used to discriminate the NLOS multipath. We implement two supervised learning methods, a support vector machine (SVM) and a neural network (NN), and compare their performance. In addition, we also propose an automated method of collecting training data for LOS and NLOS signals of machine learning. The evaluation of the proposed NLOS detection method in an urban environment confirmed that NN was better than SVM, and 97.7% of NLOS signals were correctly discriminated.

Keywords: GNSS; GPS; SDR; machine learning; multipath; signal classification.