Accessibility statement

Cognitive Networking for Heterogeneous Wireless Systems

The purpose of this project is to examine how cognitive networking techniques can be applied to systems of heterogeneous wireless nodes (especially those delivering multicast transmissions), as a way of improving dynamic performance over multi-hop backhaul networks.

Traditionally wireless ad hoc networking has focussed on the development of routing protocols and metrics for homogeneous terrestrial nodes.  Recently cognitive networking has started to be investigated, where nodes intelligently route and reroute traffic in response to changes in the wireless environment.  The purpose of this project is to examine how cognitive networking techniques can be applied to systems of heterogeneous wireless nodes (especially those delivering multicast transmissions), as a way of improving dynamic performance over multi-hop backhaul networks. 

It is expected that networks will include terrestrial, high altitude platform, and satellite nodes.  A main area of interest will be to understand how different routing metrics can operate with such diverse node types, with the aim of controlling QoS, and maximising network scalability and reliability. 

A mixture of simulation and analysis will be used to assess performance.  Set theory and queuing theory will be important analytical tools.  It is intended that this work will integrate closely with the other work on cognitive networking underway within the Group.

Key objectives

  • To develop routing metrics that will operate successfully in a heterogeneous node system, which will cognitively respond to changing requirements.
  • To implement a simulation and/or analytical model of the system to quantify performance relative to more conventional shortest path routing.
  • To understand how network topology design will impact on performance.

Outputs

  • Routing metrics for heterogeneous wireless networks.
  • Simulation and/or analysis showing the relative routing performance for different heterogeneous topologies.
  • Contributions to conference and journal papers.

Members

  • Mohamed Elsokkary
  • David Grace

Dates

  • October 2008 to
    September 2009

Research