Multi-agent AI for software-defined traffic control in satellite networks

About the Project

The specific focus of this PhD research project is on integration of non-terrestrial networks (NTN) (aka satellite communication networks) with terrestrial networks (TN) (5G/6G mobile networks). The high level of industry investment in the past decade in low-earth orbit (LEO) satellite systems has formed mega satellite constellations, which can offer satellite communication capabilities with high data rates across the globe. Integrating LEO constellations with terrestrial 5G/6G networks has great potential for the betterment of the society by providing true global network coverage to people and businesses.

This research work will tackle a set of technical challenges that the communication systems industry is facing today. The capabilities of modern artificial intelligence (AI) and machine learning (ML) techniques have high potential in addressing these challenges towards effective NTN-TN integrations.

As a PhD student, you will specifically study the use of "AI agents" that can self-govern their respective domains of responsibility in an integrated networking system. The AI agents will receive performance and status metrics from the network, employ learning algorithms, and take actions to manage network control via software-defined networking (SDN) APIs. By means of forming a collaborative multi-agent environment, the system will dynamically modify packet routing policies for reliable content delivery and uninterrupted communication sessions in an integrated satellite and mobile network setting.

The output from this research will make significant contributions to University of Surrey's substantial efforts to further strengthen its leadership in satellite communications. The student will have access to the world-class test bed research facilities offered by Surrey’s flagship research institutes: Institute for Communication Systems (ICS) and Surrey Space Institute (SSI). The student will also have the opportunity to work with partner organisations in the industry, and deploy their solutions in ICS’s 6G data centre for experimental evaluations.

Supervisors: Dr Serdar Vural and Professor Rahim Tafazolli

Entry requirements

Open to candidates who pay UK/home rate fees. See UKCISA for further information. Starting in October 2026. Later start dates may be possible, please contact Dr Serdar Vural once the deadline passes.

You will need to meet the minimum entry requirement for our PhD programme.

Candidate Profile

The ideal candidate for this PhD studentship will demonstrate good skills, motivation, and a passion for software development for telecommunication networks with a research engineering mindset. You must have a strong background in computer networks and software programming, preferably in Python and/or Go. Experience in computer networks in past course work, personal projects, or participation in hackathons is highly valued. Experience and knowledge in software-defined networks (SDN), artificial intelligence (AI), machine learning (ML), cloud computing technologies (e.g. Kubernetes and Docker environments), and software development lifecycle management with Git are highly advantageous and desired. Prior experience and project work in AI/ML are specifically relevant. The student is expected to have an interest in mathematical modelling, network protocols, and systems design. The research output will be frequently disseminated in international conferences and journals, therefore the student will need to have strong technical writing skills in English. The candidate is expected collaborate with technical members of ICS and SSI and hence will need to demonstrate effective communication skills.

How to Apply

Applications should be submitted via the Information and Communication Systems PhD programme page. In place of a research proposal, you should upload a document stating the title of the project that you wish to apply for and the name of the relevant supervisor.

Funding Notes

Fully and directly funded studentship opportunity for this project only - covering home university fees, additional research training, travel funds and UKRI standard rate (£21,805 for 2026/27 academic year).

Open to candidates who pay UK/home rate fees. See UKCISA for further information.