The development of 5G mobile network technology towards the efficient and effective support of a variety of services is primarily driven by the requirements of so-called ‘verticals’, e.g. Automotive/Smart Mobility, eHealth, Media & Entertainment and Factory of the Future (FoF). Each of the verticals offers diverse services/applications with specific characteristics as well as performance requirements. For example, the FoF vertical requires 5G network technology to support both Ultra-Reliable Low-Latency Communication (URLLC) services and non-URLLC services. FoF services are generally bursty in nature, e.g. short requests for information, transfer of sensor readings or control commands.
An important 5G technology component is network slicing, where diverse services of the same or different vertical users may be served using different slices upon the same physical network infrastructure. Network slicing is supported both in the radio network and the core network. From the radio network’s perspective, cross-slice radio resource management is crucial in order to optimize the global performance of services of multiple slices.
The objective of the proposed graduation project is threefold:
- To design (one or more options for) a new adaptive radio resource management algorithm for 5G network technology supporting multiple slices, serving at least URLLC and non-URLLC services. This requires a review of the state of the art (SoTA) of random resource management algorithms for (radio) network slicing, identification of the limits of SoTA algorithms and design of new algorithms. The designed new algorithm(s) should be able to adapt to unknown or quick varying radio environments (including propagation, traffic, and network topology) via e.g. machine-learning-based methods. From the perspective of network topology, UE relaying (dynamically enabled and disabled on demand) will be considered for the sake of network coverage and availability.
- The quantitative assessment of the proposed radio resource management algorithm(s), considering a mutual comparison and a benchmark against an SoTA algorithm, in order to demonstrate attainable gains and identify the best candidate. This requires the specification of suitable and realistic scenarios in terms of system, specification of (multiple) slices, propagation and traffic aspects, and the development/application of a simulation tool to do the assessment.
- To specify one selected new radio resource management algorithm (typically in the form of a pseudo-code and execution flow diagrams) as input to the integration specification of the international CLEAR5G project (see below),as input for those other TNO colleagues who are involved in and leading the coding and integration work.
The graduation project is carried out as part of the international CLEAR5G project (www.clear5g.eu
), cooperating with partners in both the EU and in Taiwan. You will learn about the broader context of the overall project and will gain the necessary knowledge about both the ‘Factory of the Future’ vertical application domain and mobile telecommunication technology.