A GNSS Spoofing Exclusion Method Based on Density Clustering with the Rough User Position

Jianfeng Li,Hong Li,Weiyu Gao,Mingquan Lu

DOI: https://doi.org/10.33012/2020.17536

2020-01-01

Abstract:Due to the open signal structure and low power of Global Navigation Satellite System (GNSS) civil signals, GNSS receivers are vulnerable to spoofing attacks that will make the victim receiver get the wrong position, velocity, and time (PVT) results. Many anti-spoofing methods are proposed but mainly focus on spoofing detection. It is relatively difficult to identify and exclude all spoofing signals, but also more meaningful for getting the precise positioning results. In the paper, we propose a spoofing exclusion method based on density clustering by using the known rough user position. The rough position is relatively easy to get, such as through cellular network, inertial measurement unit (IMU), Loran-C, etc. We detect spoofing by comparing the difference between the rough position and the positioning result of the victim receiver. If spoofing signals exist, pseudorange-distance biases are computed by the rough position, the satellite positions, and the measured pseudoranges. Further, the density clustering algorithm for pseudorange-distance biases is used to exclude spoofing signals and recover the precise positioning results of the receiver. The detailed analyses of the factors affecting spoofing exclusion are carried out, and the relevant experimental results prove the effectiveness of the proposed method.