[Taken] Observability for Information-Centric Networks

Information-centric networking/Named-data networking provides an alternative approach to host-oriented, IP-based communication. Communication following this paradigm directs network traffic to a host that advertises a given name. A number of approaches have been implemented to demonstrate the basic viability of concepts; however, none of these approaches provide observability and introspection as provided by a lot of … Read more

[Taken] Persistent Interests for Named-Data Networking

Current communication between devices and cloud infrastructure is based largely on IPv4 or IPv6. Information-Centric Networking (ICN) proposes to retrieve information from the ‘network’ based on named content instead of focussing on communication with a endpoints identified by addresses. ICN approaches such as Named-Data Networking (NDN) require consumers to express interests for individual chunks of … Read more

[Taken] Adaptation of QUIC for Delay-Tolerant Communication

The Transmission Control Protocol (TCP) relies on frequent acknowledgements to drive the progress of transmissions. This approach works well in well-connected networks with connections that exhibit relatively low latency, however, it suffers when the connection between nodes is disrupted regularly or experiences significantly high latency. Delay-tolerant networking attempts to address this issue by introducing concepts … Read more

[Taken] AI connectivity for Information-Centric Networks

Information-centric networking assumes that content is the focus of communication and is uniquely named in networks e.g. a consumer asks a network for specific content instead of connecting to the IP address of a specific server. This approach allows content to be cached within the network and network nodes to change their location without affecting … Read more

[No longer available for 2026/27] Investigating the Impact of Data Quality

**This project is for students in the Online Statistics & Data Science programme (M.Sc) only. No other student will be accepted for this project.*   Data quality is defined as ‘the capability of data to satisfy stated and implied needs when used under specific conditions’ (International Organisation for Standardisation). Data quality issues can include missing data, misspellings and typos, … Read more

Investigating Gradient Interference and Negative Transfer in Multi-Task Deep Learning

Summary Multi-task learning (MTL) enables a single neural network to learn several related tasks simultaneously, such as image segmentation, attribute prediction, and object classification. Although MTL often improves efficiency and generalization, competing objectives from different tasks can produce conflicting gradients during training, resulting in negative transfer and degraded performance. Recent studies suggest that gradient interference … Read more

Continual Learning for Robust UAV Object Detection in Dynamic Environments

Summary: Autonomous UAVs are increasingly deployed in diverse environments such as urban areas, forests, construction sites, and agricultural fields. However, object detection models are typically trained offline using fixed datasets and often experience significant performance degradation when operating in unseen environments. Retraining the model from scratch whenever new data becomes available is computationally expensive and … Read more

Agentic AI Control of Wireless Networks Using an AI-Native Network Controller with Safety Guardrails

This project investigates the application of agentic AI reasoning to the autonomous control and optimisation of wireless communication networks, leveraging the OCUDU platform in conjunction with an AI-native network controller. The student will extend the existing controller architecture to support closed-loop agentic control, wherein an AI agent autonomously issues configuration commands, observes network responses, and … Read more

Measuring the Energy Footprint of AI Training and Inference for Network-Intelligence Workloads

1. Background and motivation As mobile networks evolve toward 6G, artificial intelligence is moving from an optional optimisation layer to the core control mechanism of the network itself: AI-native architectures increasingly rely on models embedded directly in the control loop (e.g., forecasting traffic, managing topology, orchestrating resources, and making real-time decisions across base stations and … Read more