Tharindu Lakshan Yasarathna

Converging Software-Defined Networking (SDN) and the Internet of Things (IoT) has directed innovative network architectures and applications. However, this fusion exposes security vulnerabilities due to SDN-IoT networks’ expanded attack surface and dynamic nature. Autonomous Anomaly Detection (AAD) plays a pivotal role in swiftly identifying and mitigating real-time security threats in this landscape. This survey thoroughly investigates AAD approaches within SDN-IoT networks, with a particular focus on applying advanced Deep Learning (DL) techniques. It explores the significance of SDN-IoT architecture, articulates the motivations driving AAD, and highlights its critical role in ensuring the availability, confidentiality, authentication, integrity, and authorization of networked resources. Furthermore, it defines the threat vectors across different layers of SDN-IoT, facilitating a comprehensive validation of existing attack and defence approaches. The paper provides an extensive review of DL-based AAD methods in SDN-IoT networks, examining their strengths, limitations, and practical implications. Notable contributions include an in-depth taxonomy of DL models—such as CNNs, RNNs, LSTMs, and autoencoders—designed to detect and mitigate various security threats, including DDoS attacks, data breaches, and abnormal traffic patterns. The performance of these techniques is evaluated using measures like accuracy, recall, and F1-score, with some methods achieving detection accuracy exceeding 99%. Additionally, this survey identifies key challenges in implementing AAD systems, such as scalability, real-time processing, and integration within resource-constrained IoT environments, while proposing future research directions to address these issues. In conclusion, this comprehensive survey summarizes the multifaceted landscape of DL-based AAD for Security in SDN-IoT Networks, providing a foundation for further exploration and innovation that can advance security within the evolving realm of interconnected networks.

Internet of Things (IoT) refers to the network of interconnected physical devices, vehicles, home appliances, and other items embedded with sensors, software, and connectivity, enabling them to collect and exchange data. IoT forensics aim to investigate cybercrimes, security breaches, and other malicious activities that may have taken place on these connected devices. In particular, Electromagnetic Side-Channel Analysis (EM-SCA) has become an essential tool for IoT forensics due to its ability to reveal confidential information about the internal functions of IoT devices without interfering with these devices or wiretapping their networks. However, the accuracy and reliability of EM-SCA results can be limited by device variability, environmental factors, and data collection and processing methods. In fact, very few studies have explored potential solutions to address the significant impact of these limitations on the accuracy of EM-SCA approaches applied to crossed IoT devices. Therefore, this paper examines the impact of device variability on the accuracy and reliability of machine learning (ML)-based approaches for EM-SCA. We first present the background, basic concepts and techniques used to evaluate the limitations of current EM-SCA approaches and datasets. Our study then addresses one important limitation, which is caused by the multi-core architecture of the processors (aka. System-On-Chip). We present an approach to collect the EM-SCA datasets and demonstrate the feasibility of using transfer learning to obtain more meaningful and reliable results from EM-SCA in IoT forensics of crossed-IoT devices. Moreover, this study contributes a new dataset for using deep learning (DL) models in analyzing Electromagnetic Side-Channel data with regard to the cross-device portability matter.

Clinical data's vast volume and diverse formats, compounded by the intricacies of healthcare and the heightened importance of security and privacy, necessitate sophisticated data management and governance systems. Existing studies often examine components like data distribution, privacy, and security in isolation. This research bridges the gaps by developing an efficient, integrated clinical data warehousing solution emphasizing both privacy and security at all stages. The study presents a distributed data warehouse architecture designed after meticulous research and analysis. Testing was conducted using clinical datasets, leading to the creation of privacy and security components such as the Digital Emergency Medical Card and Emergency Authenticator. The prototype's evaluations highlight a viable balance between system performance and stringent security standards, showcasing the efficiency of encryption and other security measures. Ultimately, this study offers a holistic solution, combining theory and practice, for secure clinical data management in healthcare.