Effective Detection of GNSS Spoofing Attack Using A Multi-Layer Perceptron Neural Network Classifier Trained by PSO
S. Tohidi, M. R. Mosavi
Abstract
Global Navigation Satellite System (GNSS) receivers are affected by diverse interactions from various radio frequency transmitters, either intentional or unintentional. The present work proposed a technique-based artificial Neural Network (NN) to detect spoofing attacks. This technique uses the received signal power and correlation function distortion as feature vector, and tries to classify received signals as jammed, spoofed, multi-path afflicted, or interference-free signal. In particular, a multi-layer perceptron NN trained by Particle Swarm Optimization (PSO) is proposed as a multi-classifier which is intended for classification task. To validate the performance of the proposed, the results are compared with results achieved via classification based Bayes rule. The simulation results show that spoofing attack detection has improved approximately 4% and 2% in comparison with the results achieved via classification based on Bayes-optimal rule and multi-hypothesis Bayesian classifier mentioned in literature review.