PhD Graduates
 Illustration: Part of  a communication network
See also . . .
- MSc Study
- PhD Research Study
- Network Group Seminars

Networks Research Group

The Networks Group is active in key areas in networking including: traffic modelling; quality of service and resource management; network analysis, simulation and measurement; complex and scale-free networks; security and survivability; intelligent services and ubiquitous computing; mobile and home networks.

 Notable achievements:

  • Interdisciplinary work on the application of mathematics of complexity to the behaviour of packet networks; leading to 3 granted patents. Work on network modelling and analysis using non-linear dynamical (“chaotic”) models for traffic and networks has succeeded with long-term collaboration with Mathematical Sciences, being funded by EPSRC and industry in a series of closely related projects.
  • A new topological measure to characterise the core of a network leading to a new model to describe the formation and evolution of the Internet Autonomous-System topology at a global and local level; this is considered by CAIDA (Cooperative Association for Internet Data Analysis) to be currently the most accurate Internet-topology simulator.
  • A new analysis for WiFi (supported by BT) that provides accurate modelling of the critical transition region between unsaturated and saturated conditions, with the key result that WiFi can support more VoIP traffic than previously believed.
  • New techniques in accelerated network simulation, with the Group now internationally leading that research area in both mobile and fixed networks.
  • Novel QoS approaches to ad-hoc networks that exploit node-disjoint topology.
  • Solving the underlying layer-3 issues in the exploitation of co-operative dynamic “semi-smart” antenna (patent filed) with research supported by the EU, OFCOM (with Lucent), the US Navy and the Macao Government.


  • Network Science: Build on its successful interdisciplinary collaboration (applying chaos theory and complexity) and accelerated network simulation capability to address the design of large-scale networking experiments with Prof. Gilmour (Director of QMUL Centre for Statistics) who is internationally renowned for work on the design of experiments (DOE) and surveys. DOE will be used to quantify and control errors in network measurements and optimally design network simulation experiments.
  • Ubiquitous optical interconnection: Optical Burst Switching, the crucial next step for WANs, requires new techniques to predict traffic. These techniques will be developed through interdisciplinary work with OR researchers at City and computer scientists at Royal Holloway. The Group will also exploit its position (achieved via EPSRC funding and BT Short Term Research Fellowships) through its links with Nortel Networks, BT Research Labs, research groups at DTU in Denmark and the University of Girona.
  • QoS provisioning and resource management for wireless and mobile networks: Having developed leading-edge techniques for QoS provisioning and resource management for wireless, mobile and ad-hoc networks, we will use this to produce robust and flexible systems for deployment with minimal network planning, using machine learning through exploration techniques, a crucial capability for emergency and disaster situations. The Group will establish a Sensor Network facility to exploit wireless networking knowledge in this area.
  • Home, ubiquitous and semantic networking: A critical avenue to give confidence in the developing intelligent e-society will be to exploit security through systems that can identify attacks on the applications and middleware through instrumentation of the middleware with application-domain knowledge. Funded in part by new EU project Genetic Message Oriented Secure Middleware, starting 2008 (¬£200k).