Real-time detection and optimization algorithms for flexible resource allocation in 5G networks and beyond

Abstract

The roll-out of fifth-generation (5G) mobile networks and the forthcoming sixth-generation (6G) will bring about fundamental changes in the way we communicate, access services and entertainment. With respect to the latter, the multi-fold increase in the service data rates of enhanced mobile broadband (eMBB) services will provide users with ultra high resolution in video-streaming, multi-media and virtual reality, offering immersive experiences. To this end, it is important for Edge content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. For flexible and highly adaptive solutions, the capability for quick resource (re-) allocation should be driven by early and low-complexity content popularity detection schemes. In the present thesis, we study aspects of low-complexity detection of changes in video content popularity in real-time, addressed as a statistical change point (CP) detection problem, breaking completely new ground compared to earlier works that relied upon web content research topic. Furthermore, novel exciting use cases were introduced in 5G in the context of ultra-reliable low latency communications (URLLC) and massive machine type communications (mMTC); the new industrial revolution, dubbed as Industry 4.0, along with emerging verticals in telemedicine, smart agriculture, etc., will bring about automation and intelligence to levels never seen before. As 5G is required to support a large variety of services, novel solutions to enable higher resource efficiency are sought; in this framework, in this thesis we study layer 2 scheduling of heterogeneous services, focusing in the case of URLLC and eMBB co-existence. We propose novel heuristic algorithms and further investigate solutions leveraging non-orthogonal multiple access (NOMA), because of its advantages over conventional orthogonal multiple access (OMA) schemes in terms of spectral efficiency, cell-edge throughput, and energy efficiency. All of the proposed solutions are adapted to flexible resource allocation schemes, a cornerstone of 5G and of future 6G networks.