Funding
Funded
Machine Learning Solutions for Optimising Advanced Wireless Systems Performance
School of Electronics, Electrical Engineering and Computer Science | PHD
Applications are now CLOSED
Reference Number
EEECS/2025/NS2
Application Deadline
28 February 2025
Start Date
1 October 2025
Overview
Advanced wireless systems (6G and beyond) promise to revolutionise communications by enabling seamless interconnection among distinct elements (things, people, entities, and processes—collectively known as the Internet of Everything (IoE)) within a reliable and secure environment. Traditional analytical methods struggle to meet the ambitious challenges posed by these targets, making artificial intelligence essential. Unlike previous generations, these systems will largely operate at higher frequencies to meet the rising data rate demands, where virtually every device is connected and communicating. While these frequencies allow for vast data rates, they are more prone to interference from obstacles and atmospheric conditions, causing potentially unstable communication links. Such instability is especially concerning for essential services and emergency communications. In the evolving IoE ecosystem, today’s devices are becoming more than just communication tools—they are driving network infrastructure and decision-making. As this network grows, devices will present unique data and communication needs, further complicated by dynamic user behaviours and high-frequency limitations. Traditional communication systems rely on fixed mathematical models, which struggle to adapt to these systems’ complex and dynamic environment. Machine learning (ML), however, provides the flexibility to leverage data-driven decision-making, allowing for more adaptable and reliable solutions tailored to these advanced systems’ unique requirements.
In this project, you will explore and develop novel ML-based strategies to enhance the reliability of advanced wireless systems, fostering both theoretical insights and practical applications. The project’s unique approach lies in creating tailored learning algorithms (e.g., novel loss functions) for training ML models, which incorporate reliability metrics to systematically reduce link failures. This approach seeks to address existing research gaps that have overlooked these metrics, aiming to better align model optimisation with real-world operational benchmarks and meet the high reliability requirements of advanced wireless systems’ applications.
Some key objectives include:
[1] Designing system models and strategies for channel and resource allocation, ensuring they are tailored for advanced wireless systems. These should be commercially applicable and amenable to machine intelligence. For example, a multi-user multi-selection scenario where resources are available across users, each selecting one or more at any time.
[2] Developing novel light-weight ML solutions and custom training methods for the target wireless system, aiming to improve key performance indicators, such as outage probability, error rates, ergodic capacity, and latency, while boosting communication performance.
[3] Exploring critical trade-offs such as minimising outages versus maximising network ergodic sum rate or resource selection versus transmission delay (e.g., more scanning increases delay) etc.
This research builds upon prior studies, such as [R1] which developed a custom loss function to minimise outage probability in single-user multi-resource systems, later expanded to incorporate calibration properties [R2], performance bounds [R3], and simplified approximations [R4].
With ML’s growing role in communications research, this PhD offers a unique opportunity to gain expertise in machine learning, data analysis, and wireless communications.
[R1] N. Simmons, D. E. Simmons, and M. D. Yacoub, “Outage Performance and Novel Loss Function for an ML-Assisted Resource Allocation: An Exact Analytical Framework,” IEEE Trans. on Mach. Learn. in Commun. and Netw., vol. 2, pp. 335–350, Feb. 2024.
[R2] R. Raina, N. Simmons, D. E. Simmons, and M. D. Yacoub, “Beyond Linear Binning: Logarithmic Insights for Calibrated Machine Learning in Wireless Systems,” 2024 IEEE 100th Vehicular Technology Conference, Washington DC, USA, October 2024, to be published.
[R3] R. Raina, N. Simmons, D. E. Simmons, and M. D. Yacoub, “Optimal Classifier for an ML-Assisted Resource Allocation in Wireless Communications,” in IEEE Networking Letters, vol. 6, no. 3, pp. 158-162, Sept. 2024.
[R4] R. Raina, N. Simmons, D. E. Simmons, and M. D. Yacoub, “ML-Assisted Resource Allocation Outage Probability: Simple, Closed-Form Approximations,” in 2023 IEEE Int. Conf. on Adv. Netw. and Telecommun. Syst. (ANTS), Jaipur, India, Dec. 2023, pp. 1–6.
Funding Information
To be eligible for consideration for a Home DfE or EPSRC Studentship (covering tuition fees and maintenance stipend of approx. £19,237 per annum), a candidate must satisfy all the eligibility criteria based on nationality, residency and academic qualifications.
To be classed as a Home student, candidates must meet the following criteria and the associated residency requirements:
• Be a UK National,
or • Have settled status,
or • Have pre-settled status,
or • Have indefinite leave to remain or enter the UK.
Candidates from ROI may also qualify for Home student funding.
Previous PhD study MAY make you ineligible to be considered for funding.
Please note that other terms and conditions also apply.
Please note that any available PhD studentships will be allocated on a competitive basis across a number of projects currently being advertised by the School.
A small number of international awards will be available for allocation across the School. An international award is not guaranteed to be available for this project, and competition across the School for these awards will be highly competitive.
Academic Requirements:
The minimum academic requirement for admission is normally an Upper Second Class Honours degree from a UK or ROI Higher Education provider in a relevant discipline, or an equivalent qualification acceptable to the University.
Experience with MATLAB and/or Python programming language will be beneficial.
Project Summary
Supervisor
Dr Nidhi Simmons
Mode of Study
Full-time: 3 or 3.5 years
Funding Body
Funding TBC