Resource allocation in federated learning by jointly considering communication, computation, and data heterogeneity

Eligibility: UK/International graduates with the required entry requirements

Duration: Full-Time – between three and three and a half years fixed term

Application deadline: 15 January 2025

Interview date: Will be confirmed to shortlisted candidates

Start date: May 2025

For further details contact: Abdorasoul Ghasemi

Introduction

This PhD project is part of the Cotutelle arrangement between Coventry University, UK and Deakin University, Melbourne, Australia. The PhD Student is anticipated to spend at least six months of the total period of the programme at Coventry University, with the remainder of the programme based at Deakin University. The supervision team will be drawn from the two universities.

This project will start at Deakin University, Melbourne, Australia.

Edge intelligence (EI) has been recognised as a key technology for next-generation wireless networks such as 6G. Driven by the recent success of mobile edge computing, EI pushes computation-intensive artificial intelligence (AI) tasks from the centralised cloud to distributed base stations at the wireless network edge to efficiently utilise the massive data generated by numerous edge devices. However, the large volume of data at edge devices must be properly processed (potentially jointly processed with sensed data from other sensors such as camera and Lidar) via AI algorithms in a swift manner, in order to support applications with ultralow-latency sensing, communication, computation, and control requirements. To this end, federated learning (FL) has emerged as a promising solution, where sensing devices can iteratively exchange their locally trained AI models to update the desired global AI model in a distributed manner, while preserving data privacy at each sensing device. However, given the scarcity of spectrum resources, the wireless communication for exchanging AI models between sensing devices and edge devices is recognised as the performance bottleneck for FL. To tackle this issue, over-the-air FL (Air-FL) has recently been proposed, which allows distributed edge devices to concurrently transmit their local gradient or model updates for “one-shot” aggregation at each communication round. In this way, communication and computation are seamlessly integrated to enhance the communication efficiency in FL, thereby supporting various low-latency and communication-efficient applications.

Project details

In this project, we aim to optimally manage the communication and computation resources in Air-FL, by considering the effects of data heterogeneity. The selected candidate will be expected to:

Conduct a literature review on federated learning, edge computing, resource allocation, and mathematics behind optimisation.
Design an effective resource allocation scheme between sensing and communication, when communication and sensing orthogonally coexist.
Develop optimal client sampling strategies that tackle both resource and data heterogeneity to minimise the training time with convergence guarantee in Air-FL systems with resource-constrained edge devices.
Conduct theoretical and experimental evaluations of the proposed techniques.

Funding

Tuition fees and stipend

Benefits

The successful candidate will receive comprehensive research training including technical, personal and professional skills. All researchers at Coventry University (from PhD to Professor) are part of the Doctoral College and Centre for Research Capability and Development, which provides support with high-quality training and career development activities.

This is an exciting opportunity to study a PhD as part of a cotutelle arrangement between Coventry University, UK and Deakin University, Melbourne, Australia. The PhD Student will graduate with two PhDs, one from Deakin University and one from Coventry University, each of which recognises that the program was carried out as part of a jointly supervised doctoral program.

Candidate specification

Applicants must meet the admission and scholarship criteria for both Coventry University and Deakin University for entry to the cotutelle programme.

Applicants should have graduated within the top 15% of their undergraduate cohort. This might include a high 2:1 in a relevant discipline/subject area with a minimum 70% mark (80% for Australian graduates) in the project element or equivalent with a minimum 70% overall module average (80% for Australian graduates).
A Bachelor's degree in a relevant field requiring at least four years of full-time study, and which normally includes a research component which is equivalent to at least 25% of a year’s full-time study in the fourth year, with achievement of a grade for the project equivalent to a H1 standard or 80%.

a Masters degree, with a significant research component, in a relevant subject area, with overall mark at minimum Distinction.
In addition, the mark for the Masters thesis (or equivalent) must be a minimum of 80%.
Please note that where a candidate has 70-79% and can provide evidence of research experience to meet equivalency to the minimum first-class honours equivalent (80%+) additional evidence can be submitted and may include independently peer-reviewed publications, research-related awards or prizes and/or professional reports.
Language proficiency (IELTS overall minimum score of 7.0 with a minimum of 6.5 in each component).

The potential to engage in innovative research and to complete the PhD within a prescribed period of study.

Additional Requirements

Please consider the requirements below when preparing your Expression of Interest (EoI). Please spend half the EoI outlining why your experience makes you a good fit for the project and the other half identifying and discussing some relevant literature.

The selected candidate should/be:

Have backgrounds in computer science, information technology, electronic engineering, or related areas.
Knowledgeable in machine learning, edge computing, or wireless communications (had successful courses or projects)
Proficient in programming (preferably in Python)
Ideally familiar with optimisation

How to apply

To find out more about the about the technical details of the project, please contact: Professor Yong Xiang.

In the first instance, please submit your expression of interest (EOI) via the button below with a supporting statement detailing your suitability with evidence of the following. Please spend half the EOL outlining why your experience makes you a good fit for the project and the other half identifying and discussing some relevant literature.

Have backgrounds in computer science, information technology, electronic engineering, or related areas.
Knowledgeable in machine learning, edge computing, or wireless communications (had successful courses or projects)
Proficient in programming (preferably in Python)
Ideally familiar with optimisation

Apply to Coventry University

a man leans over a laptop, wearing earbuds