RF-based drone detection and identification using deep learning approaches: An initiative towards a large open source drone database

The deployment of Unmanned Aerial Vehicles (UAVs) has gained increasing interest in various applications during recent years due to its multiple features. At the same time, the misuse of UAVs has also posed serious concerns related to privacy, safety, and security of individuals and organizations. This has caused the utmost demand for drone surveillance systems to monitor suspicious activities of any UAVs. In this context, this paper aims at presenting a systematic literature review of the most interesting works concerning the design and development of drone surveillance systems for target detection and tracking based upon RF/WiFi techniques. More specifically, the research tackles the addressed design issues and challenges, the employed methods and techniques. Furthermore, the methodology of the literature review is carefully described, and the performance assessment of relevant studies is explored and analyzed, based on specific criteria. Provided the analysis of state-of-the-art, future research trends are also highlighted in detail, in order to drive the scientific community to research and design novel and efficient drone surveillance systems based on RF/WiFi techniques.

Since the past few years, the commercial unmanned aerial vehicle (UAV) sector has grown significantly, thus making easier access of UAVs to public. Because these gadgets have the potential to inflict major risks intentionally or accidentally, this development has quickly aroused security concerns. Since then, many researchers have been working to develop a counter-drone system using different options such as acoustics, RF analyzers, optics, etc. This paper presents an idea on developing and implementing detection model of uncooperative quadcopter drones and track them simultaneously by using a 2-degree-of-freedom robotic manipulator with an onboard camera. First, the detection model is discussed which is developed by training a modified You look only once (YOLO v2) object detector with a large dataset of labeled images of quadcopter. Further, the target's depth is estimated via the stereo camera integrated with RGB. Tracking of target drone is done by controlling the manipulator in Pulse Width Modulation (PWM) mode in which signal is sent as soon as the target is detected.

As a new type of aerial robotics, drones are easy to use and inexpensive, which has facilitated their acquisition by individuals and organizations. This unequivocal and widespread presence of amateur drones may cause many dangers, such as privacy breaches by reaching sensitive locations of authorities and individuals. In this paper, we summarize the performance-affecting factors and major obstacles to drone use and provide a brief background of deep learning. Then, we summarize the types of UAVs and the related unethical behaviors, safety, privacy, and cybersecurity concerns. Then, we present a comprehensive literature review of current drone detection methods based on deep learning. This area of research has arisen in the last two decades because of the rapid advancement of commercial and recreational drones and their combined risk to the safety of airspace. Various deep learning algorithms and their frameworks with respect to the techniques used to detect drones and their areas of applications are also discussed. Drone detection techniques are classified into four categories: visual, radar, acoustics, and radio frequency-based approaches. The findings of this study prove that deep learning-based detection and classification of drones looks promising despite several challenges. Finally, we provide some recommendations to meet future expectations.