The adoption of dynamic spectrum sharing requires addressing various technical challenges, such as intelligent sensing and radio frequency (RF) spectrum awareness. In this scenario, we present a new framework that utilises variational Bayes factor analysis (VBFA) for cooperative wideband spectrum sensing (WSS) without any prior assumption. The approach is data-driven and capable of detecting unused spectrum bands using a test statistic on an evidence lower bound (ELBO). The framework is then applied to a case study that considers shadowing effects and frequency-selective multipath channels between primary users (PUs) and sensors. The solution performs better than the state-of-the-art methods, demonstrating excellent performance in low signal-to-noise ratio (SNR) environments, e.g., reaching a detection probability of 90% for a nominal SNR of $-10 \mathrm{d B}$.

In this work, we propose a novel framework for reactive jammer detection based on spectrum patrolling, blind source separation (BSS), and causal inference. The methodology is based on a spectrum patrol that performs jamming detection without demodulating the received signals; therefore, it can manage scenarios with legitimate users belonging to different networks and adopting completely different network protocols. Furthermore, because of the wireless medium, over-the-air signals captured by the patrols are mixed; consequently, BSS is used to separate traffic patterns. The proposed causal inference-based approach (namely all-versus-one transfer entropy (AvOTE)) reaches a remarkably high probability of detection (i.e., 0.95) with low false alarm probability (i.e., 0.05) even in high shadowing regimes. Moreover, to fully understand the impact of different parameters on the performance, we study the impact of the signal-to-jammer ratio (SJR) and the jammer attack strategy on the detection performance in different shadowing regimes.

Jamming attacks to hinder communication capabilities are becoming a critical aspect of wireless networks. A challenging issue is the detection of reactive jammers that perform spectrum sensing and attack the network only when legitimate communication is in progress. In this scenario, we introduce a novel framework for reactive jamming detection using a patrol of radio-frequency (RF) sensors external to the network to be protected. The solution relies on two key components: i) a novel underdetermined blind source separation (UBSS) method that, starting from the signal mixtures observed by the RF patrollers, is capable of separating the jamming temporal profile from the network nodes’ transmission profiles; ii) a new jamming detection based on causal inference called all-versus-one transfer entropy (AvOTE). The framework is then applied to a case study where the victim network is a Long Range (LoRa)-based internet of things (IoT) system with star topology. The solution outperforms a state-of-the-art method and an approach that attempts to find the causal relationship via time series correlation, exhibiting very good performance in the presence of shadowing. Indeed, a detection probability of 90% is achieved with a false alarm probability of 6% in the presence of nuisances such as collisions and severe shadowing.