http://bura.brunel.ac.uk/handle/2438/8622 Wed, 17 Jun 2026 08:56:35 GMT 2026-06-17T08:56:35Z http://bura.brunel.ac.uk/handle/2438/33398 Title: RGCNet: Riemannian graph convolutional networks for end-to-end smart contract vulnerability detection Authors: Chen, Y; Zhu, H; Li, H; Yang, Y; Wang, Q; Li, M Abstract: Frequent security issues with smart contract vulnerabilities have become a pressing challenge in the industry. Conventional program analysis methods lack flexibility and extensibility, leading to high false positive rates. Deep learning approaches are emerging as a new trend to address this issue. Compared to other neural networks, graph convolutional networks can better capture the structural and logical information of smart contracts. However, existing methods do not fully consider the scale-free characteristics of smart contracts and fail to leverage their complex hierarchical structures and semantic information. Therefore, we develop an end-to-end vulnerability detection framework using Riemannian Graph Convolutional Networks (RGCNet). We first construct smart contract graphs that are rich in semantic and structural information. Next, we learn features of the smart contract graph in the Riemannian manifold, thereby better reflecting its actual topology. Simultaneously, the word embedding network extracts semantic features, forming an end-to-end network where modules promote one another. Extensive experiments are conducted on three vulnerabilities using real-world smart contracts. The results show that the proposed approach exhibits superior performance over state-of-the-art methodologies in terms of accuracy, precision, and recall. Description: Data availability: Data will be made available on request. Thu, 21 May 2026 00:00:00 GMT http://bura.brunel.ac.uk/handle/2438/33398 2026-05-21T00:00:00Z http://bura.brunel.ac.uk/handle/2438/33397 Title: A Preface to the Special Issue: Emerging Areas in Network and Intelligence Empowered Computing Authors: Li, M; Wang, P Abstract: The past decade has seen networking and intelligent computing converge into a single discipline, where connectivity is no longer a passive substrate but an active, learning system. Ubiquitous IoT devices, cloud–edge continuums, and programmable data planes now generate torrents of telemetry that demand real-time inference, closed-loop control, and autonomy. From 5G/6G network slicing to intent-based and self-organizing networks, the stack is becoming software-defined, data-driven, and increasingly adaptive – setting the stage for AI models that reason over graphs, streams, and spatiotemporal patterns, and for networks that optimize themselves under tight latency, reliability, and energy constraints. Mon, 27 Oct 2025 00:00:00 GMT http://bura.brunel.ac.uk/handle/2438/33397 2025-10-27T00:00:00Z http://bura.brunel.ac.uk/handle/2438/33390 Title: Blockchain Security Solution for a Dynamic Edge Computing Platform for Enhanced IIoT Application Performance Authors: Al-Mubarakis, HH; Jedidi, A; Al-Raweshidy, H Abstract: In the context of the Internet of Things (IoT), the combined use of edge computing and blockchain technology is rapidly expanding. This approach improves performance, energy efficiency, storage management, and most importantly, security and privacy within IoT environments. IoT devices are also economically valuable across a range of industries, such as healthcare and manufacturing. However, because of their limited resources, energy constraints, and heightened vulnerability to intrusions, IoT devices and especially mobile IoT devices that move between networks face significant security challenges. Traditional security methods often become ineffective as the volume of data generated by mobile IoT devices continues to grow, which means that users become increasingly exposed to risks. To address these challenges, we propose a new method that integrates blockchain, edge computing, and mobile IoT devices. This method is known as the Secure Edge Mobility Protocol (SEMP) and is designed specifically for mobile IoT devices operating within edge computing environments. In the SEMP framework, IoT devices are able to roam smoothly and securely between networks through the dynamic generation of secure cryptographic keys created using blockchain technology and based on a modified PoAh protocol. This key generation mechanism strengthens device security while also improving system performance and energy consumption. The proposed SEMP architecture uses a decentralised structure, strong consensus mechanisms, and advanced encryption techniques that are suitable for resource-constrained mobile IoT devices. SEMP maintains high standards of privacy and performance for industrial IoT applications. It is implemented and rigorously tested under a wide range of realistic conditions. The results show significant improvements in security, robustness, scalability, and energy efficiency. Tue, 07 Apr 2026 00:00:00 GMT http://bura.brunel.ac.uk/handle/2438/33390 2026-04-07T00:00:00Z http://bura.brunel.ac.uk/handle/2438/33386 Title: Accurate Interference and Coverage Modelling in Finite OWC Networks Authors: Mahbas, A; Cosmas, J; Nilavalan, N; Al-Raweshidy, H Abstract: Accurate modelling of interference and coverage in optical wireless communication (OWC) systems remains challenging due to the limitations of conventional approaches, which typically rely on infinite-network assumptions or simplified disc-shaped cell models. In practical deployments, OWC networks are finite and regularly structured, resulting in spatially varying interference patterns that are not captured by existing models. This paper proposes a comprehensive analytical framework for evaluating interference and coverage probability in finite OWC networks with regularly deployed grid-based nodes. The framework is developed for a baseline line-of-sight (LOS)-dominant scenario with regularly spaced nodes, ideal transmitter–receiver alignment, and unobstructed propagation conditions (i.e., without blockage or misalignment effects). It explicitly accounts for three-dimensional distances, inter-node spacing, system dimensions, and transmitter–receiver height differences, while incorporating boundary effects. To capture spatial variability, the network is partitioned into core, mid, and boundary zones. Semi-analytical expressions for the interference distribution are derived for each zone, revealing distinct behaviours and pronounced performance degradation in cell-edge regions. Analytical and simulation results demonstrate that commonly adopted disc-assumption models significantly overestimate system performance by neglecting edge effects. For example, at a signal-to-interference-plus-noise ratio (SINR) threshold of −3 dB, disc-based models predict approximately 95% coverage, whereas the proposed framework and simulations show that only about 75% of the core and mid zones satisfy this threshold. The results further show that increasing inter-node distance and adopting higher reuse factors substantially improve coverage, while larger height differences degrade performance by increasing the number of visible interferers. Overall, the proposed framework provides a realistic and generalisable tool for analysing finite OWC networks, enabling more accurate performance evaluation and more reliable network design and deployment. Tue, 02 Jun 2026 00:00:00 GMT http://bura.brunel.ac.uk/handle/2438/33386 2026-06-02T00:00:00Z