The steady growth in the number of deployed Internet of Things (IoT) devices has been paralleled with an equal growth in the number of malicious software (malware) targeting those devices. In this work, we build a detection mechanism of IoT malware utilizing Control Flow Graphs (CFGs). To motivate for our detection mechanism, we contrast the underlying characteristics of IoT malware to other types of malware—Android malware, which are also Linux-based—across multiple features. The preliminary analyses reveal that the Android malware have high density, strong closeness and betweenness, and a larger number of nodes. We show that IoT malware samples have a large number of edges despite a smaller number of nodes, which demonstrate a richer flow structure and higher complexity. We utilize those various characterizing features as a modality to build a highly effective deep learning-based detection model to detect IoT malware. To test our model, we use CFGs of about 6,000 malware and benign IoT disassembled samples, and show a detection accuracy of ≈99.66%.

Cloud computing is an innovative technique that offers shared resources for stock cache and server management. Cloud computing saves time and monitoring costs for any organization and turns technological solutions for large-scale systems into server-to-service frameworks. However, just like any other technology, cloud computing opens up many forms of security threats and problems. In this work, we focus on discussing different cloud models and cloud services, respectively. Next, we discuss the security trends in the cloud models. Taking these security trends into account, we move to security problems, including data breaches, data confidentiality, data access controllability, authentication, inadequate diligence, phishing, key exposure, auditing, privacy preservability, and cloud-assisted IoT applications. We then propose security attacks and countermeasures specifically for the different cloud models based on the security trends and problems. In the end, we pinpoint some of the futuristic directions and implications relevant to the security of cloud models. The future directions will help researchers in academia and industry work toward cloud computing security.

IoT malware detection using control flow graph (CFG)-based features and deep learning networks are widely explored. The main goal of this study is to investigate the robustness of such models against adversarial learning. We designed two approaches to craft adversarial IoT software: off-the-shelf methods and Graph Embedding and Augmentation (GEA) method. In the off-the-shelf adversarial learning attack methods, we examine eight different adversarial learning methods to force the model to misclassification. The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Intensive experiments are conducted to evaluate the performance of the proposed method, showing that off-the-shelf adversarial attack methods are able to achieve a misclassification rate of 100%. In addition, we observed that the GEA approach is able to misclassify all IoT malware samples as benign. The findings of this work highlight the essential need for more robust detection tools against adversarial learning, including features that are not easy to manipulate, unlike CFG-based features. The implications of the study are quite broad, since the approach challenged in this work is widely used for other applications using graphs.

The acceptance of the Internet of Things (IoT) for both household and industrial applications is accompanied by the rapid growth of IoT malware. With the increase of their attack surface, analyzing, understanding, and detecting IoT malicious behavior are crucial. Traditionally, machine and deep learning-based approaches are used for malware detection and behavioral understanding. However, recent research has shown the susceptibility of those approaches to adversarial attacks by introducing noise to the feature space. In this work, we introduce DL-FHMC, a fine-grained hierarchical learning approach for robust IoT malware detection. DL-FHMC utilizes Control Flow Graph (CFG)-based behavioral patterns for adversarial IoT malicious software detection. In particular, we extract a comprehensive list of behavioral patterns from a large dataset of malicious IoT binaries, represented by the shared execution flows, and use them as a modality for malicious behavior detection. Leveraging machine learning and subgraph isomorphism matching algorithms, DL-FHMC provides state-of-the-art performance in detecting malware samples and adversarial examples (AEs). We first highlight the caveats of CFG-based IoT malware detection systems, showing the adversarial capabilities in generating practical functionality-preserving AEs with reduced overhead using Graph Embedding and Augmentation (GEA) techniques. We then introduce Suspicious Behavior Detector, a component that extracts comprehensive behavioral patterns from three popular IoT malicious families, Gafgyt, Mirai, and Tsunami, for AEs detection with high accuracy. The proposed detector operates as a model-independent standalone module, with no prior assumptions of the adversarial attacks nor their configurations.

In smart grids (SGs), the systematic utilization of consumer energy data while maintaining its privacy is of paramount importance. This research addresses this problem by energy theft detection while preserving the privacy of client data. In particular, this research identifies centralized models as more accurate in predicting energy theft in SGs but with no or significantly less data protection. Current research proposes a novel federated learning (FL) framework, namely FedDP, to tackle this issue. The proposed framework enables various clients to benefit from on-device prediction with very little communication overhead and to learn from the experience of other clients with the help of a central server (CS). Furthermore, for the accurate identification of energy theft, the use of a novel federated voting classifier (FVC) is proposed. FVC uses the majority voting-based consensus of traditional machine learning (ML) classifiers namely, random forests (RF), k-nearest neighbors (KNN), and bagging classifiers (BG). To the best of our knowledge, conventional ML classifiers have never been used in a federated manner for energy theft detection in SGs. Finally, substantial experiments are performed on the real-world energy consumption dataset. Results illustrate that the proposed model can accurately and efficiently detect energy theft in SGs while guaranteeing the security of client data.

Internet of Drones (IoD) offers several advantages, such as enabling ground communications in areas with connectivity limitations due to physical obstacles. However, IoD also introduces new security vulnerabilities. For instance, an attacker could launch a drone capture attack to obtain confidential credentials and inject counterfeit information into the IoD network, jeopardizing its smooth operation. In such scenarios, authentication plays a crucial role in ensuring security. Unfortunately, the existing authentication frameworks are insufficient to protect against attacks in the IoD environment. In this article, we propose a secure and anonymous authentication framework called SAAF-IoD for the IoD. SAAF-IoD utilizes a chaotic map in combination with symmetric encryption and hash functions. It allows external users of the IoD network to establish a session key with the deployed drone by performing mutual authentication with the Ground-Station Server (GS). The proposed SAAF-IoD framework provides resistance against privileged insider attacks and GS bypass attacks. We validate the resilience of SAAF-IoD against various security attacks through both formal and informal analyses. Furthermore, we compare the efficiency of the proposed framework with a benchmark scheme in terms of communication and computational costs. Our results demonstrate that SAAF-IoD requires fewer resources than the relevant schemes while offering enhanced security features.