This paper presents a federated learning-based approach for detecting threats in edge-cloud networks. The system adapts to new anomalies with dynamic signature models, maintaining low latency and high accuracy without centralized data aggregation, ensuring privacy-preserving real-time cybersecurity monitoring.

Oliver Bradley McAllister, Fiona Isabelle Redgrave, Jack William Forsythe, Emily Kate Sanderson, Rupert Francis Holloway

This research introduces an AI-based patching strategy that forecasts system vulnerabilities and automates distribution in cloud ecosystems. The model integrates security analytics and threat intelligence to reduce exploit windows, offering proactive remediation in infrastructure-as-a-service (IaaS) environments.

Clara Evangeline Bowden, Tobias Arthur Whitmore, Julian Henry Bexley, Sophie Helena Mallory, Edward James Truscott

This study proposes a behavioral access control framework for decentralized cloud platforms. Risk levels are computed using context-aware metrics like user location, device fingerprint, and behavior history to authorize or restrict access in real time, improving security resilience.

Patrick Adrian Lonsdale, Florence Margaret Atherton, Daniel Charles Moorcroft, Rebecca Alice Fielding, Thomas Gerald Quenby

This paper develops a quantum key distribution protocol tailored for multi-cloud communications. Using entangled photon pairs and verification algorithms, the system secures data channels against MITM attacks, promoting robust inter-cloud cryptographic resilience in post-quantum threat scenarios.

Samuel Dominic Hargreaves, Imogen Elizabeth Carver, Oscar Elliot Drayton, Phoebe Harriet Linwood, Henry Maxwell Uxbridge

This paper introduces a hybrid model that combines statistical threat detection with AI predictions to adjust load balancing strategies for mission-critical cloud services. It improves both performance and resilience by prioritizing traffic through least-vulnerable nodes during cyber incidents.

Lucy Isabelle Fenwick, Alexander George Crowther, Chloe Beatrice Langdon, Jasper William Fitzroy, Felicity Anne Downing

This study introduces a temporal attack correlation model for forecasting cyber threats in cloud infrastructure. By analyzing time-sequenced threat data, the system identifies future intrusion trends and dynamically triggers security configurations to prevent breaches before actual compromise.

Frederick John Cartwright, Penelope Grace Ellsworth, Hugo Nathaniel Redman, Matilda Claire Barrington, Edward Lionel Godfrey

This paper proposes a lightweight encryption protocol to safeguard virtual machine migration between distributed cloud data centers. The model minimizes performance overhead while ensuring confidentiality and integrity, enabling seamless secure mobility of workloads in multi-tenant architectures.

Julian Michael Etheridge, Harriet Victoria Langley, Rowan David Northbridge, Alice Frances Merton, Dominic Charles Wetherby

This study develops a decentralized identity verification system using blockchain for secure multi-cloud access. It eliminates centralized failure points and enhances transparency, allowing users to control authentication without relying on third-party providers, improving trust and system independence.

Charles Edmund Bramley, Victoria Eleanor Millington, Sebastian Hugo Bramford, Olivia Mary Sutcliffe, Benjamin George Tennant

This paper introduces a graph-embedding-based context analysis framework for identifying insider threats in federated cloud systems. Behavioral context from access logs is modeled using node embeddings to highlight anomalies, enhancing proactive detection of unauthorized actions within organizations.

Isabelle Florence Huxley, Nathaniel Oscar Linton, Louisa Rachel Althorp, Thomas William Grantham, Clara Madeleine Fenner