PhD Studentship: Reliable Resource Orchestration for Agile Aerial Edge Computing-PhD

Mobile Edge Computing (MEC) has emerged as a promising computing paradigm to support emerging high-performance applications by deploying resources at the network edge. However, most existing MEC systems are built upon fixed ground infrastructure, such as terrestrial base stations. This stationary deployment lacks the necessary flexibility to adapt to dynamic environments. In scenarios where ground communication is interrupted, such as during disaster recovery, in remote industrial sites, or for temporary large-scale events, fixed infrastructure is either unavailable or easily damaged, creating "blind spots" in service coverage. To overcome these limitations, Aerial Edge Computing (AEC) is required to provide an agile, on-demand computing layer.

By mounting edge servers on Unmanned Aerial Vehicles (UAVs), AEC enables flexible deployment and can establish Line-of-Sight (LoS) communication links with ground users, significantly improving signal quality. This mobility allows AEC nodes to "follow" the demand, providing high-speed data transfer and low-latency processing exactly where and when it is needed. Despite its potential, AEC faces unique challenges for Quality-of-Service (QoS) guarantees due to the highly dynamic mobility of UAVs, limited onboard energy, and the stochastic nature of 3D wireless channels. Therefore, this project aims to develop novel reliable resource orchestration solutions for AEC by harvesting recent breakthroughs in Machine Learning (ML) and analytical modelling. Specifically, this project seeks to quantify key performance metrics and create powerful adaptive ML-driven management methods to control risk while maximizing resource utilization and minimizing energy consumption.

WP1: QoS Quantitative Analysis (Months 1-11): This WP focuses on developing original analytical models to investigate performance behaviour and QoS metrics specifically for AEC systems. Models will analyse key features such as 3D mobility patterns, bursty traffic arrivals, and the dynamics of aerial-to-ground wireless transmissions.

WP2: Smart Resource Orchestration (Months 12-26): Driven by WP1, this WP will formulate a multi-objective optimization problem to balance latency, throughput, and the flight-related energy consumption of UAVs. A distributed ML algorithm based on Liquid State Machines will be designed to adaptively optimize resource assignment and UAV positioning.

WP3: Algorithms Validation and Use Case Demonstration (Months 26-36): Using an existing simulator, WP3 will establish an AEC testbed to evaluate the solutions from WPs 1-2. A typical use case of accurate environment perception will be developed to demonstrate AEC’s ability to support high-stakes autonomous operations.

For eligible students, the studentship will cover home tuition fees plus an annual tax-free stipend of at least £21,805 for 3.5 years full-time, or pro rata for part-time study. The student would be based in the Faculty of Environment, Science and Economy at the Streatham Campus in Exeter.

Advert information