Intelligent transport services for the future B5G/6G mobile networks

• Grant number: 22/14503-3
• Support Opportunities: Scholarships abroad - Research
• Effective date (Start): August 01, 2023
• Effective date (End): July 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Bruno Yuji Lino Kimura
• Grantee: Bruno Yuji Lino Kimura
• Host Investigator: Toktam Mahmoodi
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Federal de São Paulo (UNIFESP). Campus São José dos Campos. São José dos Campos , SP, Brazil
• Research place: King's College London, England

Abstract

The future 6G networks are expected by 2030, when 5G networks will hardly accommodate a huge traffic of thousands exabytes/month and dozen billions of mobile subscriptions. Holographic Communications, Extended Reality, Tactile Internet, Remote Control for surgeries and industry operations are examples of future applications that will demand end-to-end performance higher than what 5G can deliver, i.e., throughput above 100 Gbps and delay bellow 1 ms. To do so, research efforts have been concentrating on enabling high frequencies channels from millimetre waves and terahertz links. However, greater link error, frequent quality fluctuations, and high packet loss are expected in such high frequencies. While end-to-end performance depends on the efficiency of transport protocols, existing transport algorithms consisted of rule-based decision making. This leads such algorithms to react inefficiently to high frequency B5G/6G links, then under utilization of network resources and, hence, unsatisfying end-to-end performance. In this project, we will exploit the multi-connectivity property of the State of Art transport protocols such as QUIC and MPQUIC, specifically, modern transport services such as multi-stream and multi-path. To enable efficient end-to-end performance in the future networks, we will investigate how online learning with centralized and distributed machine-learning techniques can be applied to evolve such modern transport services. Instead of reacting to network events with inaccurate rule-based approaches, we expect to find more intelligent approaches to make accurate, proactive and efficient transport decisions. This project continues the research efforts we have initiated in the previous FAPESP project #2015/18808-0. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

CARDOSO, LUIS FERNANDO DE SOUZA; KIMURA, BRUNO YUJI LINO; ZORZAL, EZEQUIEL ROBERTO. Towards augmented and mixed reality on future mobile networks. MULTIMEDIA TOOLS AND APPLICATIONS, v. N/A, p. 36-pg., 2023-06-12. (22/14503-3)