Cooperative Multi-Node Jamming Recognition Method Based on Deep Residual Network

Abstract

Anti-jamming is the core issue of wireless communication viability in complex electromagnetic environments, where jamming recognition is the precondition and foundation of cognitive anti-jamming. In the current jamming recognition methods, the existing convolutional networks are limited by the small number of layers and the extracted feature information. Simultaneously, simple stacking of layers will lead to the disappearance of gradients and the decrease in correct recognition rate. Meanwhile, most of the jamming recognition methods use single-node methods, which are easily affected by the channel and have a low recognition rate under the low jamming-to-signal ratio (JSR). To solve these problems, a multi-node cooperative jamming recognition method based on deep residual networks was proposed in this paper, and two data fusion algorithms based on hard fusion and soft fusion for jamming recognition were designed. Simulation results show that the use of deep residual networks to replace the original shallow CNN network structure can gain a 6–14% improvement in the correct recognition rate of jamming signals, and the hard and soft fusion-based methods can significantly improve the correct jamming recognition rate by about 3–7% and 5–12%, respectively, under low JSR conditions compared with the existing single-node method.

Improved Deep Residual Shrinkage Network for Intelligent Interference Recognition with Unknown Interference

“…To evaluate the accuracy and potential of the proposed AFUCR-SNRSN model, an ablation experiment is applied to SNRSN to check each module’s performance in recognizing interference. Then, the classic SVM, 1D-CNN in [ 13 ], LRN in [ 14 ], Res-soft in [ 20 ] and SNRSN are used to compare their performance in terms of interference signal recognition on a close set. Lastly, each method above is compared with the AFUCR-SNRSN model on an open set.…”

“…In this situation, the dual problem of ( 15) can be represented as in (20), where k(x i , x j ) represents the kernel function used to project raw data to a high-dimensional space. This kernel function can be represented as ϕ(x i ), ϕ(x j ) .…”

“…Firstly, the time–frequency analysis was used to pre-process the interference signal, and then the long short-term memory network could fully obtain the characteristics of the correlation bidirectionally, extract the characteristics of the signal and finally complete the recognition. Shen [ 20 ] proposed a residual network based on soft fusion (Res-soft). Res-soft utilized multi-node cooperation to achieve data fusion, which helped to recognized interference.…”

Securing the Future: A Resourceful Jamming Detection Method Utilizing the EVM Metric for Next-Generation Communication Systems

“…Machine learning algorithms are also trained using spectrogram images [30], IQ samples [31], time-domain signal samples [32], and FFT samples [33,34] for jamming detection. Although machine learning algorithms are becoming increasingly popular, the issues of training these algorithms, collecting sufficient data for training, adapting to varying jamming strategies, and integrating them into the system architecture with minimal overhead must be considered.…”

Recognition of Micro-Motion Jamming Based on Complex-Valued Convolutional Neural Network

“…Thanks to the development of artificial intelligence technology, deep learning has been successfully applied to ISAR jamming pattern recognition [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 ]. For instance, Wang et al [ 27 ] implemented the recognition of jamming patterns by CNN for three kinds of jamming, including suppression jamming, multiple false targets jamming, and narrow-pulse jamming.…”