Defense strategies have been well studied to combat Byzantine attacks that aim to disrupt cooperative spectrum sensing by sending falsified versions of spectrum sensing data to a fusion center. However, existing studies usually assume network or attackers as passive entities, e.g., assuming the prior knowledge of attacks is known or fixed. In practice, attackers can actively adopt arbitrary behaviors and avoid pre-assumed patterns or assumptions used by defense strategies. In this paper, we revisit this security vulnerability as an adversarial machine learning problem and propose a novel learning-empowered attack framework named Learning-Evaluation-Beating (LEB) to mislead the fusion center. Based on the black-box nature of the fusion center in cooperative spectrum sensing, our new perspective is to make the adversarial use of machine learning to construct a surrogate model of the fusion center's decision model. We propose a generic algorithm to create malicious sensing data using this surrogate model. Our real-world experiments show that the LEB attack is effective to beat a wide range of existing defense strategies with an up to 82 percent of success ratio. Given the gap between the proposed LEB attack and existing defenses, we introduce a non-invasive method named as influence-limiting defense, which can coexist with existing defenses to defend against LEB attack or other similar attacks. We show that this defense is highly effective and reduces the overall disruption ratio of LEB attack by up to 80 percent.

Wireless jamming remains as one of the primary threats towards wireless security. Traditionally, jamming is able to disrupt wireless signals within, but not beyond, its covered bandwidth. In this paper, we propose a novel attack strategy, called orthogonality-sabotaging attack, against orthogonal frequency division multiple access (OFDMA) that has been widely adopted in today's wireless network standards (e.g., 4G/5G and 802.11ax). The attack intentionally introduces an unaligned narrowband jamming signal to an OFDMA network so as to destroy the orthogonality among all subcarriers in broadband signals. We theoretically formulate and optimize the attack strategies, and then use real-world experiments to show that orthogonality sabotaging is very efficient and can take down an 802.11ax network with only 1/5-1/4 of the full network bandwidth. Finally, we propose an attack identification and localization method to identify and localize orthogonality-sabotaging attacks in the fullband spectrum with 92% overall accuracy and localization errors within about 0.4 subcarrier spacing in experiments.

Traditional signal processing design (e.g., frequency offset and channel estimation) at a receiver treats each packet arrival as an independent process to facilitate decoding and interpreting packet data. In this paper, we enhance the performance of this process in the wireless network domain. We propose STAteful inter-Packet signaL procEssing (STAPLE), a framework of stateful signal processing residing between the physical and link layers. STAPLE transforms the signal processing procedure into a lightweight stateful process that caches in a small-sized memory table physical and link layer header fields as packet state information. The similarity of such information among packets serves as prior knowledge to further enhance the reliability of signal processing and thus improve the wireless network performance. We implement STAPLE on USRP X300-series devices with adapted configurations for 802.11a/b/g/n/ac and 802.15.4. The STAPLE prototype is of low processing complexity and does not change any wireless standard specification. Comprehensive experimental results show that the benefit from STAPLE is universal in various wireless networks.