2025-05-20

Title of thesis: Moving Towards Cognitive Radio Access Networks: Transforming MIMO Complexities into Opportunities

Link to thesis in Lund University Research Portal

Defence:  Thursday, June 5th, in E:1406 at 09:15.

Zoom link.
Zoom ID: 69658824268.

Describe your research in a popular science way

The introduction of advanced antenna systems marked a pivotal advancement, employing Multiple-Input Multiple-Output (MIMO) technology. These systems opened new avenues for understanding channel characteristics. An apt analogy for these systems is a large astronomical observatory employing an array of telescopes scattered across the globe to observe the same distant object. Each telescope collects a unique portion of the light or radio waves, influenced by its position and angle. By combining the data from all telescopes, a more detailed and multi-dimensional image is created, surpassing what a single telescope could achieve. In recent years, modern cellular networks have faced new requirements beyond traditional communication between users and the network, including integrated wireless sensing. Wireless sensing utilizes existing communication signals to perceive and interpret the surrounding environment. From a network perspective, this involves leveraging cellular infrastructure to enable sensing without the need for additional dedicated sensors. Rather than solely transmitting data, the network can analyze signals to detect object presence and movement, user positioning, and environmental changes. My thesis partly examines various aspects of wireless sensing, a field still in its early stages at the time of writing. The telecommunications industry is transitioning toward the next generation of networks, 6G, focusing on terabit-level speeds, holographic communication, and AI-driven networks. Unlike traditional communication paradigms, which rely on a reactive approach where receivers wait for signals, cognitive cellular networks will utilize predictive capabilities to anticipate traffic demands and upcoming events. This thesis leverages AI, which, unlike traditional programmed devices, is adaptive, probabilistic, and capable of learning from data to make autonomous decisions. The question of how AI can enable cognitive mobile networks by utilizing MIMO-based "telescopes" to predict user movement patterns or radio channel characteristics over both short and long time horizons is addressed in the thesis. AI technology holds transformative potential to revolutionize cellular network operations by leveraging historical data to enable intelligent, autonomous functionality with minimal human intervention.

Artistic illustration of the concept of spatial fingerprinting.

What made you want to pursue a PhD?

By nature, I constantly seek out new challenges. After 20 years of experience in the industry, I decided it was time to take the next step venturing into the unknown and stepping out of my comfort zone. It’s in these moments, when the ground feels uncertain, that I tend to grow the most as a person.

What is the most fascinating or interesting with your thesis subject?
 
How the interaction between technological challenges and recent advances in machine learning and AI offers new insights and solutions to complex problems.

Do you believe some results from your research will be applied in practice eventually? And if so, how / how?

Definitely, when I began my PhD, this research field was still largely unexplored, especially from an industrial perspective. Today, AI is regarded as essential for the future of telecommunications systems.

What are your plans?

As an industrial PhD at Ericsson in Lund, it feels natural to return and apply the skills and insights I’ve gained.