Recent years have seen a substantial increase in interest in deepfakes, a fast-developing field at the nexus of artificial intelligence and multimedia. These artificial media creations, made possible by deep learning algorithms, allow for the manipulation and creation of digital content that is extremely realistic and challenging to identify from authentic content. Deepfakes can be used for entertainment, education, and research; however, they pose a range of significant problems across various domains, such as misinformation, political manipulation, propaganda, reputational damage, and fraud. This survey paper provides a general understanding of deepfakes and their creation; it also presents an overview of state-of-the-art detection techniques, existing datasets curated for deepfake research, as well as associated challenges and future research trends. By synthesizing existing knowledge and research, this survey aims to facilitate further advancements in deepfake detection and mitigation strategies, ultimately fostering a safer and more trustworthy digital environment.

With the rapid advancements and notable achievements across various application domains, Machine Learning (ML) has become a vital element within the Internet of Things (IoT) ecosystem. Among these use cases is IoT security, where numerous systems are deployed to identify or thwart attacks, including intrusion detection systems (IDSs), malware detection systems (MDSs), and device identification systems (DISs). Machine Learning-based (ML-based) IoT security systems can fulfill several security objectives, including detecting attacks, authenticating users before they gain access to the system, and categorizing suspicious activities. Nevertheless, ML faces numerous challenges, such as those resulting from the emergence of adversarial attacks crafted to mislead classifiers. This paper provides a comprehensive review of the body of knowledge about adversarial attacks and defense mechanisms, with a particular focus on three prominent IoT security systems: IDSs, MDSs, and DISs. The paper starts by establishing a taxonomy of adversarial attacks within the context of IoT. Then, various methodologies employed in the generation of adversarial attacks are described and classified within a two-dimensional framework. Additionally, we describe existing countermeasures for enhancing IoT security against adversarial attacks. Finally, we explore the most recent literature on the vulnerability of three ML-based IoT security systems to adversarial attacks.

Spectrum sensing aims at searching and finding the unused frequency bands in specific radio spectrum. It monitors the frequency bands to detect the activity of primary/licensed users and decide if secondary users can use these bands or not. In order to improve the efficiency of spectrum sensing in wideband cognitive radio networks, compressive sensing framework has been recommended and studied in many papers since it helps the system to get better and faster results using the sparse structure of the radio spectrum. Therefore, this paper represents an in-depth survey of the best requirements of compressive sensing and spectrum sensing techniques for robust combination and effective solution for wideband cognitive radio networks. It also provides examples of innovative applications of compressive spectrum sensing including IoT, smart city and 5th generation of mobile networks. To sum up some challenges and research directions related to compressive spectrum sensing technique are given at the end.

Cooperative network is a promising concept for achieving a high-accuracy decision of spectrum sensing in cognitive radio networks. It enables a collaborative exchange of the sensing measurements among the network users to monitor the primary spectrum occupancy. However, the presence of malicious users leads to harmful interferences in the system by transmitting incorrect local sensing observations.To overcome this security related problem and to improve the accuracy decision of spectrum sensing in cooperative cognitive radio networks, we proposed a new approach based on two machine learning solutions. For the first solution, a new stacking model-based malicious users detection is proposed, using two innovative techniques, including chaotic compressive sensing technique-based authentication for feature extraction with a minimum of measurements and an ensemble machine learning technique for users classification. For the second solution, a novel deep learning technique is proposed, using scalogram images as inputs for the primary user spectrum’s classification. The simulation results show the high efficiency of both proposed solutions, where the accuracy of the new stacking model reaches 97% in the presence of 50% of malicious users, while the new scalogram technique-based spectrum sensing is fast and achieves a high probability of detection with a lower number of epochs and a low probability of false alarm.

Coronavirus (COVID-19) is continuing its spread across the world, with more than seven million confirmed cases. The findings could be important as lockdown restrictions begin to be eased, and they highlight the need for the introduction of increasingly effective techniques to deal with this spread and help effectively identify new infections more quickly, at a reasonable cost and with a minimum error rate. The use of machine learning models constitutes a new approach, used more and more in this field. In this proposed work, we built a new classification model named CovStacknet and it based on StackNet metamodeling methodology combined with deep convolutional neural network as the basis for features extraction from XRay images. The proposed model has reached an accuracy score of 98%, which is better than that achieved by the basic models.