Wireless communication signal generation for drone detection models in unreal engine

Drones have gained widespread popularity with different uses, including positive and malicious applications. Consequently, safeguarding territories from unauthorized drone activities has become more critical. However, the creation and testing of drone detection systems inherits challenges for some researchers due to the prohibitive costs of required equipment. In this thesis, we introduce an innovative approach by developing an Airsim GUI tool compatible with the Unreal Engine. This tool enables the generation of image and radio frequency (RF) datasets for drone detection within a simulation environment on a single screen. Our novel approach involves modeling the USRP receiver to collect RF signals as raw in-phase and quadrature (IQ) data. In addition, users of this tool can manage automated route planning, recording configurations, various cameras, and RF configurations. Researchers can now create datasets with diverse images and RF configurations without physical drones, cameras, or USRPs, facilitating experimentation with various drone detection models. Additionally, we demonstrate the use of the implemented tool across different use cases and propose drone detection and distance estimation models.