Shen Wang’s research while affiliated with University College Dublin and other places

Cloud computing has revolutionized the provisioning of computing resources, offering scalable, flexible, and on-demand services to meet the diverse requirements of modern applications. At the heart of efficient cloud operations are job scheduling and resource management, which are critical for optimizing system performance and ensuring timely and cost-effective service delivery. However, the dynamic and heterogeneous nature of cloud environments presents significant challenges for these tasks, as workloads and resource availability can fluctuate unpredictably. Traditional approaches, including heuristic and meta-heuristic algorithms, often struggle to adapt to these real-time changes due to their reliance on static models or predefined rules. Deep Reinforcement Learning (DRL) has emerged as a promising solution to these challenges by enabling systems to learn and adapt policies based on continuous observations of the environment, facilitating intelligent and responsive decision-making. This survey provides a comprehensive review of DRL-based algorithms for job scheduling and resource management in cloud computing, analyzing their methodologies, performance metrics, and practical applications. We also highlight emerging trends and future research directions, offering valuable insights into leveraging DRL to advance both job scheduling and resource management in cloud computing.

Dynamic Spectrum Sharing (DSS) is a pivotal technology for optimizing spectrum utilization and fostering efficient sharing among diverse users. However, existing DSS approaches face significant challenges related to security and privacy vulnerabilities, leading to fraudulent practices within spectrum marketplaces. In this paper, we introduce Spect-NFT, a novel framework leveraging Non Fungible Tokens (NFTs) to address these limitations and enhance the spectrum-sharing ecosystem's efficiency and revenue. Spect-NFT employs NFTs to authenticate ownership of spectrum bands, mitigating fraudulent activities and improving trust among participants. Additionally, Spect-NFT introduces digital Permission Tokens (PTs) to facilitate seamless spectrum sharing between primary users (PUs) and secondary users (SUs) enabling the sharing of a single NFT among multiple owners. We present a methodology for converting spectrum licenses into NFTs and demonstrate the feasibility of our approach using the Ethereum Blockchain. Our proof of concept solution showcases Spect-NFT's tamper-resistant characteristics and its potential to revolutionize DSS paradigms.

It is vital that autonomous Unmanned Aerial Vehicles (UAVs) are able to avoid obstacles effectively. When avoiding such obstacles it is important that movement decision are made fast (i.e. inference latency is low) so that crashes are avoided. When deep reinforcement learning (DRL) is being leveraged as the method of obstacle avoidance one way of reducing this inference latency is to deploy the DRL model at the edge (e.g., on-UAV). However, even if the DRL model is small enough to be deployed on-UAV, the inference latency can be too high. There is a lack of research that examines reducing DRL inference time of UAVs when avoiding obstacles. Thus, this paper examines various model compression techniques to improve the inference speed of such DRL models deployed at the edge. We propose an approach that combines these model compression techniques and apply it to a well performing Dueling Double Deep Q-Network (D3QN) baseline model. On the Nvidia Jetson Orin Nano and Nvidia Jetson Nano edge devices we show that, relative to our baseline model, this combined model compression approach reduces inference latency by 38.61% and 53.18% while only reducing the success rate by 2.34% and 5% respectively. All our code is open-sourced.

With the advent of 5G commercialization, the need for more reliable, faster, and intelligent telecommunication systems is envisaged for the next generation beyond 5G (B5G) radio access technologies. Artificial Intelligence (AI) and Machine Learning (ML) are immensely popular in service layer applications and have been proposed as essential enablers in many aspects of 5G and beyond networks, from IoT devices and edge computing to cloud-based infrastructures. However, existing 5G ML-based security surveys tend to emphasize AI/ML model performance and accuracy more than the models' accountability and trustworthiness. In contrast, this paper explores the potential of Explainable AI (XAI) methods, which would allow stakehold-ers in 5G and beyond to inspect intelligent black-box systems used to secure next-generation networks. The goal of using XAI in the security domain of 5G and beyond is to allow the decision-making processes of ML-based security systems to be transparent and comprehensible to 5G and beyond stakeholders, making the systems accountable for automated actions. In every facet of the forthcoming B5G era, including B5G technologies such as ORAN, zero-touch network management, and end-to-end slicing, this survey emphasizes the role of XAI in them that the general users would ultimately enjoy. Furthermore, we presented the lessons from recent efforts and future research directions on top of the currently conducted projects involving XAI.

Despite its enormous economical and societal impact , lack of human-perceived control and safety is redefining the design and development of emerging AI-based technologies. New regulatory requirements mandate increased human control and oversight of AI, transforming the development practices and responsibilities of individuals interacting with AI. In this paper, we present the SPATIAL architecture, a system that augments modern applications with capabilities to gauge and monitor trustworthy properties of AI inference capabilities. To design SPATIAL, we first explore the evolution of modern system architectures and how AI components and pipelines are integrated. With this information, we then develop a proof-of-concept architecture that analyzes AI models in a human-in-the-loop manner. SPATIAL provides an AI dashboard for allowing individuals interacting with applications to obtain quantifiable insights about the AI decision process. This information is then used by human operators to comprehend possible issues that influence the performance of AI models and adjust or counter them. Through rigorous benchmarks and experiments in real-world industrial applications, we demonstrate that SPATIAL can easily augment modern applications with metrics to gauge and monitor trustworthiness, however, this in turn increases the complexity of developing and maintaining systems implementing AI. Our work highlights lessons learned and experiences from augmenting modern applications with mechanisms that support regulatory compliance of AI. In addition, we also present a road map of ongoing challenges that require attention to achieve robust trustworthy analysis of AI and greater engagement of human oversight.

In the progressive development towards 6G, the ROBUST-6G initiative aims to provide fundamental contributions to developing data-driven, AI/ML-based security solutions to meet the new concerns posed by the dynamic nature of forthcoming 6G services and networks in the future cyber-physical continuum. This aim has to be accompanied by the transversal objective of protecting AI/ML systems from security attacks and ensuring the privacy of individuals whose data are used in AI-empowered systems. ROBUST-6G will essentially investigate the security and robustness of distributed intelligence, enhancing privacy and providing transparency by leveraging explainable AI/ML (XAI). Another objective of ROBUST-6G is to promote green and sustainable AI/ML methodologies to achieve energy efficiency in 6G network design. The vision of ROBUST-6G is to optimize the computation requirements and minimize the consumed energy while providing the necessary performance for AI/ML-driven security functionalities; this will enable sustainable solutions across society while suppressing any adverse effects. This paper aims to initiate the discussion and to highlight the key goals and milestones of ROBUST-6G, which are important for investigation towards a trustworthy and secure vision for future 6G networks.

With the dawn of distributed Artificial Intelligence (AI) accelerated with the upcoming Beyond 5G (B5G)/6G networks, Federated Learning (FL) is emerging as an innovative approach to performing distributed learning in a privacy-preserved manner. Numerous techniques are available for fine-tuning AI-based parameters in FL. Depending on factors such as specifications, use cases, and limitations, this tuning method can vary among different FL processes, which is essential to consider when designing and developing FL-based systems. Therefore, in this research, we conduct an investigation on the variation and trade-offs of such AI-based parameters, with metrics including user diversity analysis using t-SNE-based projections, and we also examine AI parameter performance from a Differential Privacy (DP) perspective. In addition, we expand our study to Explainable Artificial Intelligence (XAI)-based tuning and use SHAP and LIME methods to decipher distributed model complexity. We assess the interpretability of FL models by leveraging parameters like consistency, currentness, and compacity, which provide insight into their effectiveness and resilience in practical settings.

Artificial Intelligence (AI) will play a critical role in future networks, exploiting real-time data collection for optimized utilization of network resources. However, current AI solutions predominantly emphasize model performance enhancement, engendering substantial risk when AI encounters irregularities such as adversarial attacks or unknown misbehaves due to its "black-box" decision process. Consequently, AI-driven network solutions necessitate enhanced accountability to stakeholders and robust resilience against known AI threats. This paper introduces a high-level process, integrating Explainable AI (XAI) techniques and illustrating their application across three typical use cases: encrypted network traffic classification, malware detection, and federated learning. Unlike existing task-specific qualitative approaches, the proposed process incorporates a new set of metrics, measuring model performance, explainability, security, and privacy, thus enabling users to iteratively refine their AI network solutions. The paper also elucidates future research challenges we deem critical to the actualization of trustworthy, AI-empowered networks.

The rise of mobile users, IoT devices, and data-intensive applications has led to an unprecedented surge in spectrum demand. However, current Spectrum Sharing (SS) methods, characterized by centralized control and inflexible architectures, fall short of meeting this escalating challenge. The solutions of Dynamic Spectrum Access (DSA) and Dynamic Spectrum Management (DSM) have emerged to enhance spectral efficiency and facilitate novel services in the realm of 5G networks. The successful implementation of DSM requires rapid sensing, coordination, and management, all the while upholding stringent standards for security and privacy. Unfortunately, existing DSM approaches relying on spectrum databases and Cognitive Radio (CR) techniques face issues related to reliability, security, and privacy. Blockchain (BC) emerges as a promising solution for decentralized DSM, offering superior security and privacy capabilities. Distinct features of BC, such as Smart Contracts (SCs), empower the establishment of complex Service Level Agreements (SLA) among operators. Additionally, the utilization of tokens within BC ensures a reliable and trustworthy environment for spectrum trading. Furthermore, BC's seamless integration with artificial intelligence (AI) and related Machine Learning (ML) techniques presents an opportunity to automate and enhance the adaptability of DSM frameworks. Despite the potential, there exist research gaps that warrant attention. This paper aims to comprehensively analyze BC-based DSM as the primary solution to DSM challenges and offers clear future directions for addressing BC deployment challenges.