SUPR
Resource Allocation for 5G Networks using Machine Learning
Dnr:

NAISS 2024/5-87

Type:

NAISS Medium Compute

Principal Investigator:

James Gross

Affiliation:

Kungliga Tekniska högskolan

Start Date:

2024-03-01

End Date:

2025-03-01

Primary Classification:

20204: Telecommunications

Allocation

Abstract

Machine learning concepts are useful for determining a future outcome, with certain accuracy, based on the presently available information. The input data set for machine learning can be analysed in different ways, depending upon which learning algorithm is used and what are its relevant parameter settings. Currently, extensive research on fifth generation or 5G systems is going on, where the goal is the provision of high data rates in medium-to-high mobility scenarios. One possibility is using the machine learning algorithms to predict the possible allocation of resources in a 5G system to a user in a particular scenario. In this way, optimal system design can be achieved without the need for on-spot computation, which will reduce the computational overhead.