Machine Learning Approaches for Resource Allocation in Sixth Generation (6G) Wireless Systems

It is anticipated that machine learning and artificial intelligence (AI) will find places among dominant technologies in 6G wireless communication systems. The core activities where AI will play crucial role include self-configuration, dynamic topology optimization, energy and spectrum management, security and authentication, position detection and prediction for moving elements such as unmanned areal vehicles (UAVs) and autonomous vehicles, channel estimation, and link selection for enabling ultra low latency connections.

Deep learning is becoming increasingly popular choice for enabling AI at various levels. The inherent property of deep neural network (DNN) that will make it a driving force for 6G and massive IoT is the ability to find solutions for the extremely large dimensional problems. Thus, deep learning will play significant role in distributed optimization for ultra dense wireless networks in which classical centralized and distributed optimization approaches can no longer cope with the scale and heterogeneity of the network. However, the DNN framework itself needs to be designed in a way that it can handle the distributed nature and heterogeneity of 6G network infrastructure. Machine learning and AI will guide new security solutions as well by predicting behavioural models of suspicious activities of malicious nodes. Machine learning techniques will also facilitate cross-layer security design, localize potential eavesdroppers as well as make physical layer security more practical with real-time channel data.

This project theme aims to have a comprehensive overview of the existing resource allocation techniques, in terms of their capabilities and limitations in 6G, followed by developing machine learning techniques for resource allocation in different scenarios. Topics of interest to be covered by this project theme include, but are not limited
to:
• Optimal resource allocation in 6G
• Low-latency communications
• Deep learning based channel estimation
• Vehicle-to-vehicle (V2V) communications
• Unmanned aerial vehicle (UAV) communications
• Physical layer security

Knowledge in wireless communications and Matlab are required for the successful completion of this project theme. The student should be willing to closely collaborate with other students and work in a team that aims to develop realistic 6G technologies.

Supervisor name: 
Dr Muhammad Khandaker
Supervisor and Deputy email addresses: 
m.khandaker@hw.ac.uk

Project Type:

Project location: 
Earl Mountbatten Building