Smart Jamming: Deep Learning-Based UAVs Neutralization System

Recently the proliferation of unmanned aerial vehicles (UAVs) has emerged as a pressing concern, attributed to the accessibility of capable drones and the surge in incidents involving unauthorized UAVs flying over restricted areas, such as airports, stadiums, military bases, etc. Consequently, the development and deployment of drone jamming techniques have been proposed to address this evolving threat, particularly by law enforcement agencies. However, these jamming technologies require high transmitted power and risk interference with other communications operating within the same industrial, scientific, and medical (ISM) band. This paper introduces a smart jamming technique to neutralize UAVs threats. The proposed approach uses radio frequency (RF) analysis and deep learning techniques to identify the drone model and generate tailored jamming signals to minimize overall power consumption to neutralize the target drone. Simulation results show that the proposed approach outperforms other jamming techniques. The proposed approach attains successful jamming at low jammer-to-signal ratios (JSR) as compared to other jammers, as a result, less transmitted power is required to reach the same jamming performance as the other jamming technique.