Tutor: CESANA MATTEO
Advisor: CAPONE ANTONIO Major Research topic
:Resource management and optimization for new technologies in future wireless mobile networksAbstract:
Differently from the previous generations of mobile radio networks, the fifth generation (5G) has been designed to improve the performance of mobile communications beyond the mere data throughput, by acting on a wider range of performance indicators which include latency, reliability, connectivity and power consumption. This evolutionary paradigm is expected to support the communication needs of an ever-increasingly connected society and was made possible by the introduction of new technologies and concepts into mobile radio networks, such as new frequency bands (i.e. millimetre waves), advanced spectrum access and management techniques, virtualization, artificial intelligence and network slicing.
Additionally, the scientific community started to identify several use cases that go beyond the performance of 5G systems under development today, identifying the key technologies that will lead the long-term evolution of the current generation and shape the characteristics of future mobile radio networks.
My research proposes to apply advanced optimization, planning and learning techniques to some of the most promising technologies, aiming at exploring their application possibilities and exploiting the potential impact that they can have on the current and future generations of mobile communications.
During a preliminary investigation phase, the following technologies have been selected for study: millimetre-waves communication and integrated access and backhaul (IAB), satellite networks and space-earth integration, resource management for advanced service chains in network slicing and intelligent surfaces.
Thanks to the large portion of available spectrum, millimetre-waves (mmWaves) communications promises unprecedented throughput performance. One interesting application of this technology is the same-band integration of backhaul and access transmissions in dense small cell networks, with a substantial reduction in network deployment costs. However, the poor propagation characteristics of such high frequencies lead to communication that are easily hindered by distance, atmospheric adverse conditions and physical obstacles, which do not pose any issue to communications based on more traditional frequencies.In this context, accurate planning and optimization of network parameters such as routing, scheduling, power control and beam selection are essential for the success of this technology.
By integrating satellites with ground stations on earth, space-terrestrial networks can extend the coverage of global communication networks to levels that cannot be matched by an exclusively terrestrial network. In particular, the integration of space-based networks with ground 5G network infrastructure is gaining interest, since it can transparently offer the benefits of an integrated space-terrestrial network to mobile end users without the need to include dedicated satellite communication equipment in their mobile terminal. However, this integration process is at its early stage and different system-level challenges arise from the intrinsic characteristics of such a complex integrated network, to name a few: possibly large transmission delays, non-stationary satellites, intermittent rain attenuation, limited capacity and upgradeability of satellites.
5G communications are expected to reach and revolutionize segments of the industry which are not currently exploiting the full potential of mobile services. The newly introduced concept of Network Slicing addresses this issue by letting Infrastructure Providers (InP) to operate different and multiple logical networks (slices) on a single physical infrastructure. A high level of slice customization allows the tenant (owner of a slice) to tune the performances and service characteristics of its slice to the requirements of its business operations. Furthermore, newly itroduced Mobile Edge Computing (MEC) offers computing capabilities closer to the radio access segment, enabling advanced service chains in which data is processed and transferred at the same time. In this resource sharing context, efficient resource management is fundamental in assuring that the slicing system works as intended, while being both efficient and fair.
Most recently, reconfigurable intelligent surfaces (RIS) with particular electromagnetic properties are gaining popularity as a disruptive technology for future wireless networks. RIS are usually defined as planar structures made of several individually tuneable elements that can be programmed to control the characteristics of incoming electromagnetic signals, ultimately reflecting or refracting them towards specified locations. One or more RIS applied to objects create a so-called smart radio environment, a candidate architecture that has the potential to dramatically increase both efficiency and efficacy of future wireless communication by sensing the environment and applying customized transformation to the radio waves. However, this exceptionally promising technology is in its very early stage and large research efforts are needed to leverage smart radios environments as a unified platform that integrates communications, sensing, and computing.