Implementation of authentication scheme for GNSS signal using Raspberry Pi

• Main Author: Shi, Wenpu
• Other Authors: Tan Soon Yim
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/160992
• Institution: Nanyang Technological University

GNSS (Global Navigation Satellite System) could provide time synchronization and measurement of object position services.The GNSS receiver can calculate its current position through the navigation information transmitted by the satel- lite. However, all transmissions of civilian GNSS signals are unprotected. Ad- ditionally, the receiver needs to receive signals from multiple navigation satel- lites in different directions. The receiving antenna beam is very wide, almost like an omnidirectional antenna, which can receive radio signals from all di- rections around and regarded them all as GNSS signals. Combined with the power of the GNSS signal when it reaches the earth’s surface is extremely low, which makes spoofing the GNSS receiver quite simple. The GNSS spoofing is refers to the attacker send the fake GNSS signals to make the receiver generate wrong position, time, velocity solutions. Some non cryptography methods that detect the energy and Doppler consistency of the received signals to detecting the GNSS spoofing were proposed to protect the GNSS signals, however these methods would be inaccurate due to the environment influences. Yu Han Chu et. al had proposed a navigation message authentication scheme which used the unbounded chameleon hash keychain to provide protection to GNSS signals. In this system only the dedicated authentication server was allowed to created the new keys to extend the keychain which has an unique properties-unbound, and any clients could connected to this system after authenticating. The server will send the digital signature (HMACs) of each subframes of the navigation mes- sages along with the collision pair to clients after receive the client’s request. The clients will first use received the collision pair to authenticate whether the response is from the authentication server and then determine the authenticity of the received subframes by comparing their reference HMACs with the HMACs provided by the authentication server. The reason of the success rate of Yu Han’s authentication system model achieved by using Android phones is very low may be the authentication server is acted by phone which is unusable. Hence, this thesis is based on implementing this authentication system in an improved version by using the Raspberry PI, as a proof of concept to demonstrate the feasibility of this authentication scheme on IOT (internet of things) devices and increase the success rate. We have made sets of experiments on the open area, thus there are enough experiments data to prove that this authentication system could not only be stable connection but also correctly labeled the received signal in about practically 99.5% practised success rate. We also used the HACK-RF to to generate spoofed GPS signals to test the performance of the system, the result was exciting that reached prac- tically 100% detecting rate. Generally, this improved authentication system has been fully proven to have excellent anti-spoofing capabilities and can be widely used in most IOT devices. Additionally, in the future, more experiments will be done in different situations such as in the sheltered places (car park, shopping mall), and some bad weathers, to test the performance of this authentication system.