Graduation details

  • [PhD] - Wifi-Direct Internetworking
  • Oct 2014 - Feb 2020
  • We are on the verge of having ubiquitous connectivity. However, there are still scenarios where public communication networks are not reachable, are saturated or simply cannot be trusted. In such cases, our mobile phones can leverage device-to-device communication to reach the public network or to enable local connectivity. A device-to-device communication technology, with at least WiFi speed and range, will offer sufficient connectivity conditions for interconnection in areas/situations where it is not currently possible. Such advance will foster a new breed of systems and applications. Their widespread adoption is, nonetheless, bound to their usage in off-the-shelf devices. This raises a problem because the device-to-device communication technologies currently available in off-the-shelf mobile devices have several limitations: Bluetooth is limited in speed and range,Wi-Fi Direct is limited in speed and connectivity for medium and large scenarios, and WiFi-Aware is a new and untested technology, whose specification does not cover large scenarios. In this thesis, we address this problem by presenting two communication topologies and a network formation algorithm that enable the use of Wi-Fi Direct communication between off-the-shelf mobile devices in medium and large scale scenarios. The communication topologies, named Group-Owner Client-Relay Group-Owner and Group-Owner Group-Owner, allow for Wi-Fi Direct intergroup communication, whilst the network formation algorithm, named RedMesh, systematically creates networks of Wi-Fi Direct groups. The algorithm proved to be very effective, achieving full connectivity in 97.28% of the 1 250 tested scenarios. The RedMesh algorithm distinguishes itself for being the first one to useWi-Fi Direct communication topologies that can form tree and mesh structures, and for being the first algorithm able to build networks that can rely only on unicast communication. We may hence conclude that the work developed in this thesis makes significant progress in the formation of large scale networks of off-the-shelf mobile devices.
  • 1 Oct 2014
  • 13 Feb 2020
  • António Teófilo
  • João Lourenço, Hervé Paulino
  • José Legatheaux Martins, Luciano Baresi, Nuno Santos, Rolando Martins