Adversarial Attack and Defense on Deep Learning for Air Transportation Communication Jamming

Abstract

Air transportation communication jamming recognition model based on deep learning (DL) can quickly and accurately identify and classify communication jamming, to improve the safety and reliability of air traffic. However, due to the vulnerability of deep learning, the jamming recognition model can be easily attacked by the attacker's carefully designed adversarial examples. Although some defense methods have been proposed, they have strong pertinence to attacks. Thus, new attack methods are needed to improve the defense performance of the model. In this work, we improve the existing attack methods and propose a double level attack method. By constructing the dynamic iterative step size and analyzing the class characteristics of the signals, this method can use the adversarial losses of feature layer and decision layer to generate adversarial examples with stronger attack performance. In order to improve the robustness of the recognition model, we use adversarial examples to train the model, and transfer the knowledge learned from the model to the jamming recognition models in other wireless communication environments by transfer learning. Simulation results show that the proposed attack and defense methods have good performance.

Adversarial Attacking and Defensing Modulation Recognition With Deep Learning in Cognitive-Radio-Enabled IoT

“…(36) [39] indicates that when the model loss does not exceed the predicted loss threshold L T , the model will make a correct prediction. The predicted class of the model for an adversarial example x * is y p (x * ) = arg max…”

“…y k {p (y|x * )}, k = 1, 2, • • • , K, (37) so { y p (x * ) = y t , L (x, α, g) ≤ L T y p (x * ) ̸ = y t , L (x, α, g) > L T (38)where y p represents the predicted class of the model for the adversarial example, and y t represents the true class.APPENDIX B PROOF OF PROPOSITION 2We denote p (x * ) as p, and substitute (30) into the cross entropy loss of the model to obtainL − β) • δ k,l + β • u(k)) • log 2 (p k ) = (1 − β) [ − K ∑ k=1 δ k,l log 2 (p k ) )log 2 (p k ) ] = (1 − β) • L (x * , l) + β • L (x * , u)(39) …”

Interference Response Prediction of Receiver Based on Wavelet Transform and a Temporal Convolution Network

“…The purpose of incorporating AFF [27] is to enable adaptive amalgamation of features from various frequency components, thus guaranteeing that the fusion of features is not excessively biased towards particularly prominent traits. In certain research areas, such as radar signal type identification [28][29][30], the core idea is to identify the most unique and prominent features of the signal. However, in this research, unlike other classification tasks, the signal characteristics here are unique.…”

Spatial Parameter Identification for MIMO Systems in the Presence of Non-Gaussian Interference

“…The increasing number of wireless devices, together with the requirements for effectiveness and reliability of information transmission, have led to the development of various new communication technologies for addressing unprecedented challenges [1][2][3][4]. Among them, the multiple-input multiple-output (MIMO) technology with sensing capability has been a research hotspot [5].…”