Cluster-based STA Reporting and Multi-AP Coordination in Wi-Fi Networks

#Wi-Fi #multi-AP #coordination #channel #selection
Share

Due to the few available Wi-Fi channels and the existence of other technologies, the channel selected by an access point (AP) to be assigned to the network is of extreme importance for network performance due to different interference conditions at different channels. While channel selection algorithms have been proposed, very few of them are user-centric, which could incur a large signaling overhead. As a result, in this thesis, a channel selection algorithm with a low-overhead STA reporting mechanism is proposed, which makes use of the spatial correlation of interference by clustering close-by STAs in order to reduce the feedback overhead by adopting cluster-based feedback instead of STA-based. As a result, the channel selection feedback overhead could be reduced without sacrificing the accuracy of the feedback.

In addition, the lack of coordination between APs exacerbates performance degradation due to interference. As a result, multi-AP coordination has become inevitable. Multiple studies have proposed multi-AP coordination schemes which aim to achieve coordination in either the spatial or the frequency domain. Despite the improvement shown by current multi-AP coordination schemes, their uni-dimensional design places a limitation on the ability of the network to coordinate the transmissions sufficiently in varying scenarios where the domain for coordination is not guaranteed to remain suitable. As a result, this thesis proposes a multi-AP coordination scheme with a multi-dimensional design that aims to manage interference by allowing the APs to concurrently deploy coordination on both the spatial and frequency domains, thus providing a higher degree of reliability.



  Date and Time

  Location

  Hosts

  Registration



  • Add_To_Calendar_icon Add Event to Calendar

Loading virtual attendance info...

  • Contact Event Host
  • Starts 11 April 2023 01:00 PM UTC
  • Ends 15 April 2023 02:00 AM UTC
  • No Admission Charge