We propose to study how to build an all-wireless Internet that could grow and gradually replace the existing Internet. In our vision, the access points of unlicensed cellular networks cooperate to create a wireless communication infrastructure. Such a wireless network must be self-organizing to maximize the network capacity and capabilities. Ultimately, the all-wireless interconnection of cooperating access points will only require limited connectivity to the wired infrastructure. We call this all-wireless infrastructure the Radio Internet. The goal of the project is to study how to make the Radio Internet become a reality with an Internet-like growth towards a scalable and robust communication infrastructure. The objective of the project is to design, implement, and experimentally validate an all-wireless interconnection architecture based on advanced wireless transmission techniques (i.e., new MAC, new wireless standards, multiple radio, and multiple antennas APs), mesh networking, cross-layer optimization, and mechanisms for seamless mobility. We will develop a suitable measurement methodology and set up experimental prototypes to validate our concepts. Testbeds will be used or deployed for proof-of-concept and validation of the proposed architecture and communication mechanisms. To achieve the vision of the Radio Internet, we need to adopt a disruptive approach with respect to the current Internet: rethink the overall architecture with wireless networks as the central technology and propose a new interconnection architecture founded on the principle of all-wireless networks with some isolated wired long-haul links. By defining new topology and routing schemes, designing simple interconnection rules, elaborating new performances objectives, designing self-configuring and self healing mechanisms, a large number of wireless access networks can be dynamically interconnected to provide wide coverage with a large capacity. In addition, such a network infrastructure will enable new mobility perspectives and new business models.

Partners: Intel, CTCT, Siemens, INPG-LSR, CERTH, Swisscom, UPMC

Project website: https://www.it.pt/Projects/Index/539

João Paulo Barraca
João Paulo Barraca
Associate Professor

My research interests include cybersecurity, distributed and virtualized systems and software.