With the decrease in the cost and size of drones in recent years, their number has also increased exponentially. As such, the concerns regarding security aspects that are raised by their presence are also becoming more serious. The necessity of designing and implementing systems that are able to detect and provide defense actions against such threats has become apparent. In this paper, we perform a survey regarding the different drone detection and defense systems that were proposed in the literature, based on different types of methods (i.e., radio frequency (RF), acoustical, optical, radar, etc.), with an emphasis on RF-based systems implemented using software-defined radio (SDR) platforms. We have followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines in order to provide a concise and thorough presentation of the current status of the subject. In the final part, we also describe our own solution that was designed and implemented in the framework of the DronEnd research project. The DronEnd system is based on RF methods and uses SDR platforms as the main hardware elements.

An energy detection (ED) algorithm, which takes into consideration three consecutive sensing events, is introduced in this letter. This algorithm, named three-event ED (3EED), takes the decision in one sensing event, considering also the event immediately before and the one immediately after it. We design the proposed algorithm to exploit the knowledge of the primary user (PU) activity duty cycle value for tracking the changes of the PU state and we prove that 3EED becomes independent of the PU model parameters for a low duty cycle. In this letter, we demonstrate that 3EED has better detection performance compared to the conventional ED (CED) algorithm, providing a correct detection probability (CDP) gain of more than 7% for a false-alarm probability (FAP) of 0.1. Also, we compare 3EED with other algorithms in terms of detection performance and complexity.

Implementation of dynamic spectrum access (DSA) in cognitive radio (CR) systems requires the unlicensed secondary users (SU) to implement spectrum sensing to monitor the activity of the licensed primary users (PU). Energy detection (ED) is one of the most widely used methods for spectrum sensing in CR systems, and in this paper we present a novel ED algorithm with an adaptive sensing threshold. The three-event ED (3EED) algorithm for spectrum sensing is considered for which an accurate approximation of the optimal decision threshold that minimizes the decision error probability (DEP) is found using Newton’s method with forced convergence in one iteration. The proposed algorithm is analyzed and illustrated with numerical results obtained from simulations that closely match the theoretical results and show that it outperforms the conventional ED (CED) algorithm for spectrum sensing.

Spectrum sensing (SS) based on energy detection (ED) is a simple yet effective approach to detect the presence of unknown signals that are active in a specific band of frequencies. The classical ED (CED) algorithm uses the value of the energy detected in the current sensing slot as test statistic and has a fixed threshold, but improved signal detection performance is possible by modifying the test statistic and/or the detection threshold. In this paper we propose a novel ED algorithm for SS that considers a binary activity model for the signal to be detected and combines the use of an average energy test statistic with an adaptive decision threshold for improved detection performance. We present the analytical characterization of the proposed test statistic in terms of its mean and variance, and derive the expressions corresponding to the correct decision probability (CDP) and false alarm probability (FAP). Using the derived CDP and FAP expressions, we also determine the detection thresholds that yield desired values for the false alarm and missed detection probabilities. The proposed algorithm is illustrated with numerical results obtained from simulations, which confirm our theoretical findings and also show that the algorithm outperforms alternative adaptive ED algorithms.