Research on UAV Jamming Signal Generation Based on Intelligent Jamming

With the widespread application of UAV technology, various countries are facing new threats in both military and civilian domains. To effectively counter these threats, UAV communication jamming has become a reliable countermeasure. Traditional communication jamming methods face several challenges. These include low efficiency, difficulties in managing dynamic and complex electromagnetic environments, and a heavy reliance on prior knowledge about the signals. This study presents intelligent jamming techniques for UAV signal processing to overcome the limitations found in current communication jamming technologies. The aim is to achieve precise jamming of individual target signals using intelligent jamming methods. This paper introduces an enhanced jamming signal generation algorithm built on the traditional convolutional autoencoder. Without relying on prior signal knowledge, the algorithm introduces complex convolutional networks and residual modules. The model leverages advanced convolutional networks to integrate the real and imaginary components of the signal for feature extraction, while the residual modules further refine signal processing and enhance feature learning. The model generates jamming signals by altering hidden layer features. These signals closely resemble the original communication waveforms. Simulation results show that, across different communication systems, the generated jamming signal waveforms exhibit strong similarity to the original signal waveforms. Moreover, under various signal-to-interference ratios (SIR), the jamming performance of the generated signals surpasses that of conventional jamming signals such as Gaussian noise and amplitude modulation noise.