Advanced AI-empowered Secure, Green, and Robust 6G networks

Description

With the rapid development of wireless technologies, it is expected that future telecommunication systems can serve an enormous number of heterogeneous wireless devices with high-rate and low-latency communications. However, they face critical challenges that need to be addressed to ensure their successful deployment and operation. 

Specifically, wireless communication is highly vulnerable to radio attacks, particularly eavesdropping attacks. Using cryptographic techniques is a common countermeasure, but the heterogeneity of future networks poses challenges to the key distribution and management. Moreover, with quantum computing, encryption techniques will soon become defenceless. Physical-layer security solutions, such as friendly jamming and beamforming, have also been proposed; however, they require prior information about eavesdroppers, which is often unavailable. 

Moreover, the gigantic increase in network traffic and the deployment of a massive number of cells, edge computing nodes, and technologies like massive MIMO and THz communication, will add computational burdens to systems, resulting in extremely high energy consumption. To meet sustainability goals, one potential approach is adjusting the transmit power of BSs based on users’ demands. Nevertheless, optimising the power control policy for BSs serving diverse and highly dynamic users is challenging. 

In addition, it is expected that AI will enable zero-touch management in 6G networks. However, current AI approaches usually require massive training data and long training time. This may prolong end-to-end communication latency, increase energy consumption, and hinder 6G applications. Moreover, traditional AI approaches are trained on specific environments/scenarios, making them less effective in new system conditions. Unfortunately, this is likely the case in future communication systems where networks are highly dynamic. 

This PhD project aims to address the above challenges by 

1. Developing a novel approach to exploit co-channel interference when reusing resource blocks to disrupt the signal reception at eavesdroppers without requiring their prior information. 

1. Developing a federated deep reinforcement learning-based approach to jointly optimise user association, resource allocation, and power control while considering security, energy efficiency, and communication resilience of the system. 

1. Developing quantum-inspired neural networks to reduce the number of training parameters and energy consumption while quickly adapting to new conditions and maintaining good performance. 

Throughout the program, the student will be closely supervised by both lead supervisors and receive comprehensive training on AI, 5G/6G, and physical layer security. Transferable skills training will be also provided. The student is expected to submit their thesis to UoL at the end of Year 4. 

We want all of our staff and Students to feel that Liverpool is an inclusive and welcoming environment that actively celebrates and encourages diversity. We are committed to working with students to make all reasonable project adaptations including supporting those with caring responsibilities, disabilities or other personal circumstances. For example, if you have a disability you may be entitled to a Disabled Students Allowance on top of your studentship to help cover the costs of any additional support that a person studying for a doctorate might need as a result. 

We believe everyone deserves an excellent education and encourage students from all backgrounds and personal circumstances to apply. 

Informal enquiries are encouraged before a formal application and should be directed to Dr Huynh Nguyen(huynh.nguyen@liverpool.ac.uk) and Dr Te-Chuan Chiu (theochiu@cs.nthu.edu.tw). Candidates wishing to apply should complete the University of Liverpool application form [How to apply for a PhD - University of Liverpool] applying for a PhD in Electrical Engineering and Electronics.