Ericsson is one of the largest providers of telecommunications equipment in the world. By collaborating with our customers, we have access to large amounts of live telecom network data. With our Cloud RAN product customers have new options for how to deploy their telecom networks, among other things this gives the opportunity for resource pooling by e.g., centralizing baseband compute.
The goal of this thesis is to develop a model for predicting the impact of pooling compute resources, in terms of energy efficiency, within different regions of a telecom network. For this thesis you will work with real world data collected from live networks.
The following steps are envisioned as part of the thesis work:
- Design and develop a machine learning based solution for predicting energy efficiency of resource pooling.
- Evaluate and visualize model performance.
The thesis will be concluded with a result presentation for relevant teams at Ericsson.
This project aims at students in electrical engineering, computer science, computer engineering or similar. A background in artificial intelligence and machine learning is preferred.
1-2 students, 30hp each
Ericsson AB Mjärdevi, Linköping
Preferred Starting Date
AI/ML, data science, big data
Fredrik Sjöstrand Finn Sjögren
+46 761 17 30 83 +46 730 43 58 76