GPS Spoofing Detection by Neural Network Machine Learning

Bose, Sam C.

doi:10.1109/maes.2021.3100844

Cited by 12 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These approaches utilize features extracted from standard GNSS receivers' acquisition, tracking, and Position-Velocity-Time (PVT) modules as input for training various supervised ML models in classification tasks. For instance, Bose et al [25] employed an FCNN that incorporated features such as C/N 0 , pseudorange, carrier phase, and Doppler shift to identify spoofing attempts. Similarly, Aissou et al [9] used a set of 13 features extracted from tracking and PVT solution blocks to train multiple ML classifiers.…”

Section: A Related Workmentioning

confidence: 99%

“…However, it's worth noting that some works, such as [9], [11], [25], [31], incorporate PVT solutions as features. However, this inclusion can introduce additional detection delays since PVT solutions are acquired over a look-back sample window.…”

Section: A Related Workmentioning

confidence: 99%

“…To ensure robustness in the evaluation, we repeated the performance assessment 10 times with different train-test splits, presenting the average classification results in Table 3. In addition to the five models trained on our featureset, we included three other models; FCNN-S [26], FCNN-B [25], and Nu-SVM [9]-for comparative analysis. These models were trained using the featuresets introduced in their respective papers.…”

Section: ) Performance Under Supervised Learningmentioning

confidence: 99%

See 2 more Smart Citations

A Deep Learning Based Induced GNSS Spoof Detection Framework

Iqbal,

Aman,

Sikdar

2024

Trans. Mach. Learn. Comm. Netw.

View full text Add to dashboard Cite

The Global Navigation Satellite System (GNSS) plays a crucial role in critical infrastructure by delivering precise timing and positional data. Nonetheless, the civilian segment of the GNSS remains susceptible to various spoofing attacks, necessitating robust detection mechanisms. The ability to deter such attacks significantly enhances the reliability and security of systems utilizing GNSS technology. Supervised Machine Learning (ML) techniques have shown promise in spoof detection. However, their effectiveness hinges on training data encompassing all possible attack scenarios, rendering them vulnerable to novel attack vectors. To address this limitation, we explore representation learning-based methods. These methods can be trained with a single data class and subsequently applied to classify test samples as either belonging to the training class or not. In this context, we introduce a GNSS spoof detection model comprising a Variational AutoEncoder (VAE) and a Generative Adversarial Network (GAN). The composite model is designed to efficiently learn the class distribution of the training data. The features used for training are extracted from the radio frequency and tracking modules of a standard GNSS receiver. To train our model, we leverage the Texas Spoofing Test Battery (TEXBAT) datasets. Our trained model yields three distinct detectors capable of effectively identifying spoofed signals. The detection performance across simpler to intermediate datasets for these detectors reaches approximately 99%, demonstrating their robustness. In the case of subtle attack scenario represented by DS-7, our approach achieves an approximate detection rate of 95%. In contrast, under supervised learning, the best detection score for DS-7 remains limited to 44.1%.

show abstract

Section: A Related Workmentioning

confidence: 99%

Section: A Related Workmentioning

confidence: 99%

Section: ) Performance Under Supervised Learningmentioning

confidence: 99%

See 1 more Smart Citation

A Deep Learning Based Induced GNSS Spoof Detection Framework

Iqbal,

Aman,

Sikdar

2024

Trans. Mach. Learn. Comm. Netw.

View full text Add to dashboard Cite

show abstract

“…However, given the peculiarities of the power supply of low-orbit satellites, we can conclude that the intruder satellite does not have enough energy coming from the solar panels to effectively block the signal transmission through the radio channel. Therefore, the most likely destructive impact on the LOSIS is the imposition of relay spoofing interference [54,55]. To impose a relay spoofing interference, an intruder satellite must do the following.…”

mentioning

confidence: 99%

Improved Spacecraft Authentication Method for Satellite Internet System Using Residue Codes

et al. 2023

View full text Add to dashboard Cite

Low-orbit satellite internet (LOSI) expands the scope of the Industrial Internet of Things (IIoT) in the oil and gas industry (OGI) to include areas of the Far North. However, due to the large length of the communication channel, the number of threats and attacks increases. A special place among them is occupied by relay spoofing interference. In this case, an intruder satellite intercepts the control signal coming from the satellite (SC), delays it, and then imposes it on the receiver located on the unattended OGI object. This can lead to a disruption of the facility and even cause an environmental disaster. To prevent a spoofing attack, a satellite authentication method has been developed that uses a zero-knowledge authentication protocol (ZKAP). These protocols have high cryptographic strength without the use of encryption. However, they have a significant drawback. This is their low authentication speed, which is caused by calculations over a large module Q (128 bits or more). It is possible to reduce the time of determining the status of an SC by switching to parallel computing. To solve this problem, the paper proposes to use residue codes (RC). Addition, subtraction, and multiplication operations are performed in parallel in RC. Thus, a correct choice of a set of modules of RC allows for providing an operating range of calculations not less than the number Q. Therefore, the development of a spacecraft authentication method for the satellite internet system using RC that allows for reducing the authentication time is an urgent task.

show abstract