A New Threat for Pseudorange-Based RAIM: Adversarial Attacks on GNSS Positioning

Yuan Sun,Li Fu

doi:10.1109/access.2019.2939141

Abstract

Global Navigation Satellite System (GNSS) signals are very vulnerable to spoofing due to the low power level and opening service mode. Although pseudorange-based Receiver Autonomous Integrity Monitoring (RAIM) method performances effectively in spoofing detection and exclusion through a consistency check method, it might still suffer from threats which are deliberately designed. Besides an exhaustive research on the defense method, a deep understanding of the possible attack modes of the spoofing methods is also crucial for positioning security. This paper proposes a new threat for pseudorange-based RAIM, named adversarial attacks on GNSS positioning, which can mislead the victim receiver deducing any targeted position within a certain range without triggering the RAIM alarm. First, the adversarial attack is formulated as an optimization problem that minimizes the difference between the misled GNSS positioning results and the targeted position with RAIM missing detection. Then, the Fast Gradient Sign Method (FGSM) is applied to solve the optimization problem and obtain the adversarial examples for GNSS spoofing. Simulated and real GNSS data experiments are designed to verify the threat range and efficiency of the adversarial attacks on GNSS positioning.

Highlights

In the past few decades, with the rapid development of satellite navigation systems, e.g. Global Positioning System (GPS), Galileo, BeiDou Navigation Satellite System (BDS) and other regional systems, multi-constellation Global Navigation Satellite System (GNSS) has been widely used in various civil and military applications
Specific fault design might mislead the victim receiver deducing any targeted position within a certain range
The purpose of this paper is to use the adversarial method to generate spoofing observations that can defeat the existing Receiver Autonomous Integrity Monitoring (RAIM) system and mislead the victim receiver deducing any targeted position within a certain range

Summary

Introduction

In the past few decades, with the rapid development of satellite navigation systems, e.g. Global Positioning System (GPS), Galileo, BeiDou Navigation Satellite System (BDS) and other regional systems, multi-constellation Global Navigation Satellite System (GNSS) has been widely used in various civil and military applications. The defense methods can be divided into Signal Processing Level (SPL), Data Bit Level (DBL), and Navigation and Position Solution Level (NPSL) according to the processing aspects of the receiver [30]. SPL always distinguishes spoofing signals and authentic GNSS signals with the utility of signal information, such as signal power [5], spatial processing [6], time of arrival [7], signal quality [8], etc. It is a very import method to protect spoofing signals from accessing to the following measurement calculating or positioning solution. Different with SPL, DBL and NPSL emphasize on measurement information, such as satellite ephemeris and pseudoranges, to detect and exclude spoofing signals that injected with counterfeit measurements [9]–[11]. NPSL is regarded as the last defense of the security system, which is crucial

Objectives

Results

Conclusion