Aluno: Marcelo Antonio Marotta
Orientador: Prof. Dr. Juergen Rochol
Título: From the Network to the Spectrum:A Software-Defined Networking (SDN)-based decision-making system for H-CRAN
Linha de Pesquisa: Redes de Computadores
Local: Sala 215 (sala de videoconferência) do Prédio 43412 do Instituto de Informática da UFRGS
– Prof. Dr. Alberto Egon Schaeffer Filho (UFRGS)
– Prof. Dr. Weverton Luis da Costa Cordeiro (UFRGS)
– Prof. Dr. José Ferreira de Rezende (UFRJ – por videoconferência)
Presidente da Banca: Prof. Dr. Juergen Rochol
Abstract: Heterogeneous Cloud Radio Access Networks (H-CRAN) incorporate Heterogeneous Networks (HetNet) and Cloud Radio Access Networks (C-RAN) concepts for next-generation cellular networks. H-CRAN exploits the heterogeneity of macro and small cells from HetNet, enabling cellular networks to achieve a higher spectral efficiency. Meanwhile, concepts from C-RAN involving baseband units and remote radio heads enable H-CRAN to insert a centralized point of processing for cellular networks, reducing capital and operational expenditures. Although H-CRAN brings several opportunities to cellular networks, it is not free from challenges. Among the different challenges existent, we highlight the most relevant ones, (a) high intercell interference; (b) critical latency constraints in long-distance wireless signal processing; and (c) poor allocation of processing resources. To address these challenges, we propose a Software-Defined Networking (SDN)-based decision-making system able to be programmatically changed to make decisions that can decrease interference, meet delay constraints, and mitigate processing underusage for long-term operation of an H-CRAN. By adding concepts of SDN to with a simplified object-oriented API, it is possible to logically centralize the control of H-CRAN considering a pool of physically distributed equipment. The methodology employed to show the feasibility of the proposed approach is based on the development of a prototype that supports a wide range of decision algorithms. This prototype was evaluated in three case studies conducted on a simulated environment based on 3GPP specification for simulators. In the first case study, we evaluate the network overload implied by adding an SDN-based decision-making system to H-CRAN. Afterward, a decision algorithm is executed to exploit resource sharing deciding which elements of an H-CRAN can be used by two operators interchangeably to reduce interference and increase spectrum efficiency at the spectrum level. The third case study shows that the distance between cloud and remote radio heads must be considered in processing power allocation decisions and also when assigning a virtual baseband unit to the cloud. Evaluating the case studies becomes clear the need of an SDN-based decision-making system in H-CRAN to address the inherent challenges of these networks.
Keywords: Heterogeneous cloud radio access networks”, “Decision-making system”, e “Software-defined networking.