In recent years, the rapid advancement of wireless communication technologies, particularly in the development of sixth-generation networks, brought about challenges in spectrum efficiency, security, and reliability. Machine learning-based automatic modulation classification (AMC) plays a critical role in addressing these challenges by enabling efficient signal classification in dynamic environments. However, such systems remain vulnerable to adversarial attacks, which can induce machine learning-based systems into making mistakes and, by doing so, compromise applications that rely on them. Accordingly, in this study, we propose a robust AMC framework based on Bayesian neural networks (BNN) to mitigate the impact of adversarial attacks. Our approach uses a regularization term on the weight variance of the BNN to reduce the likelihood of extreme weight values, thereby enhancing model stability in adversarial settings. We also incorporate the Sinh-Arcsinh Gaussian distribution as a flexible prior to control skewness and tail behavior, thus improving the trade-off between robustness and accuracy. Experimental evaluations against common white-box adversarial attacks, such as fast gradient sign method (FGSM), projected gradient descent (PGD), and automatic PGD (Auto-PGD), demonstrate that our proposed model outperforms conventional AMC models, achieving greater resilience in low perturbation-to-noise ratio conditions. Taken together, these findings highlight the potential of Bayesian methods in developing more secure and reliable intelligent wireless communication systems.

Rapid advancements in deep learning (DL) and the availability of the large data sets have made the adoption of DL highly appealing across various fields. Wireless communication systems, including future 6G systems are anticipated to incorporate intelligent components like automatic modulation classification (AMC) for the cognitive radio and dynamic spectrum access. However, DL-based AMC models are susceptible to the adversarial attacks, which consist of crafted perturbations that aim to alternate the decision of a victim model. This study focuses on investigating and uncovering modern modulation classifiers’ vulnerability to the adversarial threats. Though attacks of this nature inherently jeopardize DL-based classifiers, contemporary attack methods typically exhibit diminished impact at the lower perturbation levels. Therefore, we introduce a novel attack approach that exploits the Riemannian manifold properties of the intricate neural networks, yielding adversarial samples with heightened efficacy at the lower perturbation powers. We thoroughly evaluate how effective various defense techniques are and demonstrate our proposed attack method’s ability to thwart them. The findings of this study shed light on the limitations and vulnerabilities of the DL-based AMC models in the face of the adversarial attacks. By addressing these challenges, we can enhance the robustness and security of these models, and pave the way for their reliable deployment in practical wireless communication systems, including the future 6G networks.

Deep learning has been adopted for a wide range of wireless communication tasks, including modulation classification, because of its great classification capability. However, deep learning models have been shown to also introduce risks and vulnerabilities. For instance, adversarial attacks craft and introduce imperceptible perturbations that compromise the accuracy of deep learning-based modulation classifiers on wireless receivers. Therefore, in this article, we propose a novel wireless receiver architecture that enhances deep learning-based modulation classifiers to defend them against adversarial attacks. Our experimental results show that our defense technique significantly diminishes the accuracy reduction that is caused by adversarial attacks by protecting modulation classifiers at least 18% more than existing defense techniques.

Fifth-generation (5G) networks provide connectivity to a massive number of devices and boost a plethora of applications in several different domains. However, the large adoption of connected devices increases attack surfaces and introduces several security threats that can severely damage physical objects and risk people’s lives. Despite existing intrusion detection systems (IDSs), there are still several challenges to be addressed in the detection of cyber-attacks. For instance, while unsupervised IDSs are required to detect zero-day attacks, they usually present high false positive rates. Moreover, most existing IDSs rely on long short-term memory (LSTM) networks to consider time-dependencies among data. However, LSTM networks have recently been shown to present several drawbacks and limitations, which put into question their performance on sequence modeling tasks. Thus, in this paper, we investigate generative adversarial networks (GANs), a promising unsupervised approach to detecting attacks by implicitly modeling systems, and alternatives to LSTM networks to consider temporal dependencies among data. We propose a novel unsupervised GAN-based IDS that uses temporal convolutional networks (TCNs) and self-attention to detect cyber-attacks. The proposed IDS leverages edge computing and is proposed for edge servers, which bring computation resources closer to end nodes. Experiment results show that our proposed IDS can be configured to satisfy different detection rate and detection time requirements. Moreover, they show that our IDS is more accurate and at least 3.8 times faster than two state-of-the-art GAN-based IDSs that are used as baselines.

Cyber-attacks cyber-physical systems (CPSs) can lead to sensing and actuation misbehavior, severe damages to physical objects, and safety risks. Machine learning algorithms have been proposed for hindering cyber-attacks on CPSs, but the absence of labeled data from novel attacks makes their detection quite challenging. In this context, generative adversarial networks (GANs) are a promising unsupervised approach to detect cyber-attacks by implicitly modeling the system. However, the detection of cyber-attacks on CPSs has strict latency requirements, since the attacks need to be stopped before the system is compromised. In this article, we propose FID-GAN, a novel fog-based, unsupervised intrusion detection system (IDS) for CPSs using GANs. The IDS is proposed for a fog architecture, which brings computation resources closer to the end nodes and thus contributes to meeting low-latency requirements. In order to achieve higher detection rates, the proposed architecture computes a reconstruction loss based on the reconstruction of data samples mapped to the latent space. Other works that follow a similar approach struggle with the time required to compute the reconstruction loss, which renders them impractical for latency constrained applications. We address this problem by training an encoder that accelerates the reconstruction loss computation. Experiments show that the proposed solution achieves higher detection rates and is at least 5.5 times faster than a baseline approach in the three studied data sets.

The presence of connected devices in homes introduces numerous threats to privacy via the analysis of the encrypted traffic these devices generate. Prior works have shown that traffic attributes such as packet size combined with machine learning techniques enable the inference of private information from Internet of Things users. One of the commonly used techniques to mitigate those privacy threats is traffic obfuscation, such as packet padding. Most padding mechanisms that were previously proposed statically select the number of bytes inserted in the packets, which incurs high overhead and ineffective privacy improvement. These static mechanisms are particularly unsuitable for networks whose traffic patterns are significantly dynamic, such as smart homes. This article proposes an adaptive packet padding approach based on software-defined networking (SDN) that adjusts the number of bytes inserted into packets in response to variations in the home network utilization. The proposed technique monitors the network to instruct a padding mechanism through a representational state transfer (REST) interface proposed in this article. This mechanism ensures that the length of packets generated by connected devices is modified. The evaluation includes four supervised learning mechanisms, random forest (RF), support vector machine (SVM), decision tree, and k-nearest neighbors (KNNs), to measure privacy improvement through the metrics accuracy, recall, and F1-score. Goodput, jitter, and packet loss induced by the proposal are also evaluated. Our proposal is shown to overcome the state-of-the-art solutions in privacy preservation with a significantly lower overhead. For instance, the accuracy of RF on identifying devices decreases from 96% to 4.96%.

In this paper, cryptographic schemes are applied to Ethernet-based layer-2 communication to provide authenticated encryption to safety- critical automotive control data. Confidentiality, integrity and authenticity are provided by combining AES with HMAC. Experimental results using low-cost hardware show that, despite the introduced cryptographic overhead, latency requirements are comfortably met for this type of communication.