Exploring the state of the art deep learning algorithms in computer visionand voice recognition for applications in RF signal processing

Drones popularity has been rapidly increasing for a wide class of applications including commercial delivery, photography, fire-fighting and environmental monitoring due to their low cost and high commercial availability as unmanned aerial vehicles. In the light of this growth, anti-drone technology has become a significant research topic of investigation due to unauthorized flying of drones in sensitive airspace where their presence is regarded as a potential threat. In this project, we will modify and use existing well-established computer vision and voice recognition deep learning models for signal classification to remotely detect the presence of rogue drones in the air. We will analyze several datasets of radio signals transmitted by popular drones in the market to extract unique fingerprints hidden in the drone signals. Deep learning models will be designed to make use of the extracted fingerprints for detecting the presence of a drone signal.

Faculty Supervisor:

Ian Frigaard

Student:

Alireza Sarraf Shirazi

Partner:

Skycope Technologies Inc

Discipline:

Engineering - mechanical

Sector:

Professional, scientific and technical services

University:

University of British Columbia