Salta al contenuto principale
Passa alla visualizzazione normale.

PIERLUIGI GALLO

Unified Radio and Network Control Across Heterogeneous Hardware Platforms

  • Autori: Kaminski, N.; Moerman, I.; Giannoulis, S.; Ruckebusch, P.; Gallo, P.; Zubow, A.; Leblon, R.; Seskar, I.; Choi, S.; De Rezende, J.
  • Anno di pubblicazione: 2016
  • Tipologia: Proceedings (TIPOLOGIA NON ATTIVA)
  • OA Link: http://hdl.handle.net/10447/209889

Abstract

Experimentation is an important step in the investigation of techniques for handling spectrum scarcity or the development of new waveforms in future wireless networks. However, it is impractical and not cost effective to construct custom platforms for each future network scenario to be investigated. This problem is addressed by defining Unified Programming Interfaces that allow common access to several platforms for experimentation-based prototyping, research, and development purposes. The design of these interfaces is driven by a diverse set of scenarios that capture the functionality relevant to future network implementations while trying to keep them as generic as possible. Herein, the definition of this set of scenarios is presented as well as the architecture for supporting experimentation-based wireless research over multiple hardware platforms. The proposed architecture for experimentation incorporates both local and global unified interfaces to control any aspect of a wireless system while being completely agnostic to the actual technology incorporated. Control is feasible from the low-level features of individual radios to the entire network stack, including hierarchical control combinations. A testbed to enable the use of the above architecture is utilized that uses a backbone network in order to be able to extract measurements and observe the overall behaviour of the system under test without imposing further communication overhead to the actual experiment. Based on the aforementioned architecture, a system is proposed that is able to support the advancement of intelligent techniques for future networks through experimentation while decoupling promising algorithms and techniques from the capabilities of a specific hardware platform.