Applications for Unmanned Aerial Vehicles (UAVs), operating in unlicensed bands, are vastly growing with the consolidation of the Internet of Things (IoT). However, those bands have become overcrowded as systems using them are continuously increasing. In this context, Cognitive Radio (CR) and spectrum sharing techniques have emerged as promising strategies to overcome the problem of spectrum scarcity in wireless networks, thus being considered as enabling technologies for the future 5G wireless networks. Thereby, the integration of CR with UAVs can bring important benefits for the massive deployment of UAVs. This paper aims to present an overview on the state of the art of CR technology for UAV communication, as well as indicate key opportunities and future challenges.

Unmanned Aerial Vehicles (UAVs) demand technologies so they can not only fly autonomously, but also communicate with base stations, flight controllers, computers, devices, or even other UAVs. Still, UAVs usually operate within unlicensed spectrum bands, competing against the increasing number of mobile devices and other wireless networks. Combining UAVs with Cognitive Radio (CR) may increase their general communication performance, thus allowing them to execute missions where the conventional UAVs face limitations. CR provides a smart wireless communication which, instead of using a transmission frequency defined in the hardware, uses software transmission. CR smartly uses free transmission channels and/or chooses them according to application’s requirements. Moreover, CR is considered a key enabler for deploying technologies that require high connectivity, such as Smart Cities, 5G, Internet of Things (IoT), and the Internet of Flying Things (IoFT). This paper presents an overview on the field of CR for UAV communications and its state-of-the-art, testbed alternatives for real data experiments, as well as specifications to build a simple and low-cost testbed, and indicates key opportunities and future challenges in the field.

Cognitive Radio technique has been recognized as one of the most promising solutions for the increasingly growing problem of spectrum scarcity in wireless networks, specially with the emerging of the Internet of Things. In cognitive radio networks, secondary users are allowed to intelligently access licensed bands of primary users, thus enhancing the spectrum utilization. In this context, for investigating the advantages of cognitive radio, Machine Learning techniques have been widely applied to predict primary users arrivals. However, the available simulators are usually complex and highly time consuming. Therefore, in this work, we propose a simple and intuitive primary user arrivals data generator, MARIO, that can produce random arrival data for multiple channels by employing Poisson process. This generator is validated by using the generated data to predict new sequences according to a Hidden Markov Model. Our results show that the data generator can be used to simulate various traffic patterns over different channels.

Cognitive Radio (CR) is considered a key enabler technology for applications that require high connectivity (e.g., Smart Cities and Internet of Things), mainly because of its spectrum sensing function. In this way, CRs can sense the spectrum environment to select the best available channel for communication and they can, potentially, use licensed spectrum bands as a Secondary User (SU). However, CR has to vacate the licensed spectrum band as soon as a Primary User (PU) intends to use the channel. In this way, one of the most challenging topics in CR is the PU arrival prediction. Therefore, this paper presents a real-data study case of PUs arrivals prediction using the Hidden Markov Model (HMM) in CR. Herein, the Energy Detector (ED) is used to detect the presence of PUs. Our results show that the traditional method of combining the ED with the HMM may not be suitable in CR networks.