Ehsan Nowroozi

The rapid growth of Internet of Things (IoT) devices and the introduction of 5G networks have created new opportunities for enhancing network services, while also introducing significant security concerns. Intrusion Detection Systems (IDS) are crucial for identifying malicious activities and unauthorized access in these environments. However, current IDS solutions face challenges such as sharing sensitive data and managing large-scale networks. Federated Learning (FL) presents a promising solution by enabling models to be trained on decentralized devices without sharing private data. This paper examines how FL can enhance IDS for IoT and 5G networks, with an emphasis on privacy and security concerns. We analyze various privacy, homomorphic encryption, and security mechanisms in FL, including Differential Privacy (DP) and secure aggregation, and their potential applications in strengthening IDS solutions. Additionally, we explore how FL contributes to the development of more secure and efficient IDS systems while addressing challenges such as data heterogeneity and security risks. Finally, we identify gaps in the existing research and propose directions for future work to enhance the robustness and practicality of FL-based IDS for IoT and 5G environments.

Machine learning and deep learning are transformative forces reshaping our networks, industries, services, and ways of life. However, the susceptibility of these intelligent systems to adversarial attacks remains a significant issue. On the one hand, recent studies have demonstrated the potential transferability of adversarial attacks across diverse models. On the other hand, existing defense mechanisms are vulnerable to advanced attacks or are often limited to certain attack types. This study proposes a random deep feature selection approach to mitigate such transferability and improve the robustness of models against adversarial manipulations. Our approach is designed to strengthen deep models against poisoning (e.g., label flipping) and exploratory (e.g., DeepFool, BIM, FGSM, I-FGSM, L-BFGS, C&W, JSMA, and PGD) attacks that are applied in both the training and testing stages, and Transfer Learning-Based Adversarial Attacks. We consider scenarios involving perfect and semi-knowledgeable attackers. The performance of our approach is evaluated through extensive experiments on the renowned UNSW-NB15 dataset, including both real-world and synthetic data, covering a wide range of modern attack behaviors and benign activities. The results indicate that our approach boosts the effectiveness of the target network to over 80% against labelflipping poisoning attacks and over 60% against all major types of exploratory attacks.

The increasing need to process large, high-dimensional datasets and the substantial computational power required have made the use of distributed cloud servers essential. These servers provide cost-effective solutions that make storage and computing accessible to ordinary users. However, they might face significant vulnerabilities, including data leakage, metadata spoofing, insecure programming interfaces, malicious insiders, and denial of service. To gain public trust in distributed computing, addressing concerns related to privacy and security while ensuring high performance and efficiency is crucial. Multiparty computation, differential privacy, trusted execution environments, and federated learning are the four major approaches developed to address these issues. This survey paper reviews and compares these four approaches based on a structured framework, by highlighting recent top-tier research papers published in prestigious journals and conferences. Particular attention is given to progress in federated learning, which trains a model across multiple devices without sharing the actual data, keeping data private and secure. The survey also highlights federated learning techniques, including secure federated learning, by detecting malicious updates and privacy-preserving federated learning via data encryption, data perturbation, and anonymization, as new paradigms for building responsible computing systems. Finally, the survey discusses future research directions for connecting academic innovations with real-world industrial applications.

Federated Learning is an approach that enables multiple devices to collectively train a shared model without sharing raw data, thereby preserving data privacy. However, federated learning systems are vulnerable to data-poisoning attacks during the training and updating stages. Three data-poisoning attacks–label flipping, feature poisoning, and VagueGAN–are tested on FL models across one out of ten clients using the CIC and UNSW datasets. For label flipping, we randomly modify labels of benign data; for feature poisoning, we alter highly influential features identified by the Random Forest technique; and for VagueGAN, we generate adversarial examples using Generative Adversarial Networks. Adversarial samples constitute a small portion of each dataset. In this study, we vary the percentages by which adversaries can modify datasets to observe their impact on the Client and Server sides. Experimental findings indicate that label flipping and VagueGAN attacks do not significantly affect server accuracy, as they are easily detectable by the Server. In contrast, feature poisoning attacks subtly undermine model performance while maintaining high accuracy and attack success rates, highlighting their subtlety and effectiveness. Therefore, feature poisoning attacks manipulate the server without causing a significant decrease in model accuracy, underscoring the vulnerability of federated learning systems to such sophisticated attacks. To mitigate these vulnerabilities, we explore a recent defensive approach known as Random Deep Feature Selection, which randomizes server features with varying sizes (e.g., 50 and 400) during training. This strategy has proven highly effective in minimizing the impact of such attacks, particularly on feature poisoning.

In the past few years, the popularity and wide use of video conferencing software enjoyed exponential growth in market size. This technology enables participants in different geographic regions to have a virtual face-to-face meeting. Additionally, it allows participants to utilize virtual backgrounds to hide their real environment with privacy concerns or to reduce distractions, particularly in professional settings. In scenarios where the users should not hide their actual locations, they may mislead other participants into assuming that the displayed virtual backgrounds are real. In this paper, we propose a new publicly-available dataset of virtual and real backgrounds in video conferencing software (e.g., Zoom, Google Meet, Microsoft Teams). The presented archive was evaluated by an exhaustive series of tests and scenarios using two well-known features extraction methods: CRSPAM1372 and six co-mat. The first verification scenario considers the case where the detector is unaware of manipulated frames (i.e., the forensically-edited frames are not part of the training set). A model trained on zoom frames that were tested with Google Meet frames can detect real background images from virtual ones in video conferencing software with 99.80% detection accuracy. Furthermore, it is possible to distinguish virtual from real backgrounds in videos created for videoconferencing software at a high detection rate of approximately 99.80%. According to our conclusions, the proposed method greatly enhanced the detection accuracy and resistance against diverse adversarial conditions, making it a reliable technique for classifying actual as opposed to virtual backgrounds in video communications. Given the described dataset provided and some preliminary experiments that we performed, we expect that it will lead to more future research in this domain.