PhD defense of Christopher Merlhe (September 16th, 2024)

Christopher Merlhe will defend his doctoral thesis, titled Resource Allocation in a Cell-less Context for 5G Wireless Networks, on Monday, September 16, 2024, at 10:00 AM at IMT Atlantique on the Rennes campus (Petit amphi).

Abstract
The significant increase in the number of users with ever-growing needs, the emergence of new services and new applications have led mobile networks to evolve. Moreover, with global warming and energy cost inflation, mobile networks must not only deliver a high throughput and quality of service to users, but also reduce their energy footprint.

The main objective of the thesis is to address these issues by studying radio resource allocation in a Cell-less context. This innovative approach redefines the paradigm of conventional cellular networks. Base stations are no longer independent but grouped into clusters. This centralizes decision-making, while crucial resource allocation information is shared between base stations, such as: user data, user channel state and scheduling information. This enables the design of new resource allocation solutions that wisely benefit from the cooperation between base stations.

The first part of the thesis studies resource allocation in conventional cellular networks by providing four contributions: an in-depth analysis of multi-user diversity, two new energy efficient scheduling solutions and a new wireless routing solution.

The second part of the thesis focuses on Cell-less context. The results of previous contributions have highlighted that Inter-Cell-Interference (ICI) management is the resource allocation problem that benefits most from Cell-less context. The combination of such approach with learnings from the first contributions, allowed the design of two new solutions that dynamically manage ICI according to the user’s channel state. In addition, this work show that the use of Joint-Transmission Coordinated MultiPoint (JT-CoMP) enhances network performance and quality of service of users.