Rudi Ball - Imperial College London

Rudi Ball
PhD / Research Assistant
Distributed Software Engineering Section
Department of Computing, Imperial College London
180 Queen's Gate, London SW7 2AZ, UK

[map]

Please note: In March 2012 I completed the PhD and joined the mobile software startup Tutot. The contact email provided is still checked periodically, but not very often. This page remains at the Department of Computing as a record of work completed during the PhD, but was last updated in June 2011.

Research

My interests are in complex mobile and distributed systems which utilise activity, mapping, message passing and inter-node communication as a means to provide distributed applications to users - most notably control systems on scaled populations of mobile devices. Applications are centered on so-called smart systems and smart cities - problems of traffic control, efficient resource usage, mobility, disaster relief and information dissemination using both centralised and distributed approaches.

Topics also include ad-hoc network simulation and mobile device programming (Android, Symbian and Windows Mobile).

I am especially interested in how the systems found in urban spaces and devices carried by people, can be utilised to provide beneficial applications and access to useful resources in the real world. A number of natural systems exist in urban spaces and these systems can be leveraged to provide useful solutions to end-users.

Supervisors

Dr Naranker Dulay | Dr Emil Lupu

Publications

Rudi Ball and Naranker Dulay. Enhancing Traffic Intersection Control with Intelligent Objects. First International Workshop the Urban Internet of Things 2010 - Programming the real-time city. Tokyo, Japan. December 2010. (paper)
Enrico Scalavino, Giovanni Russello, Rudi Ball, Vaibhav Gowadia, Emil Lupu. An Opportunistic Authority Evaluation Scheme for Data Security in Crisis Management Scenarios. 5th ACM Symposium on Information, Computer and Communications Security. April 2010. (bibtex)
Leonardo Mostarda, Rudi Ball, Naranker Dulay. Distributed Fault Tolerant Controllers. DAIS 2010: 141-154 (bibtex)
Leonardo Mostarda, Rudi Ball, Naranker Dulay. Distributed Fault Tolerant Controllers. Technical Report. TR 2010/3, Department of Computing, Imperial College London. February 2010.
Rudi Ball, Naranker Dulay. Approximating Travel Times using Opportunistic Networking. 2nd IEEE Intl Workshop on Opportunistic Networking. April, 2009 (bibtex)
Rudi Ball, James Grant, Jonathan So, Victoria Spurrett, Rogério de Lemos: Dependable and Secure Distributed Storage System for Ad Hoc Networks. ADHOC-NOW 2007: 142-152 (bibtex)

Up to date listing: DBLP

Projects

Cityware - Researching urban environments and the ways in which users utilise pervasive environments and tools.
AEDUS2 - Researching adaptive software environments; policy and security; and requirements engineering and modelling.

Simulation

A large part of my PhD work has involved work on the Geographic Urban Simulator (GUS). The GUS is a powerful discrete-event simulator for the simulation of mobile devices in urban spaces and along specified mobility tracks (traces) using collected or synthesized geographic coordinates. It has been built atop the JiST (Java in Simulation Time) framework developed by Rimon Barr at Cornell. As many as 100 000 (complex) peers have been tested on the simulator. The source-code is set for release in 2011.

Simulation of 500 commuter patterns over the city of London.

GUS simulation of 500 commuters for the city of London - traces represent the tracks taken by synthesised mobile users.

Typical message traversal visualisation.

A typical message trace generated by the simulator (message traversal between mobile devices) - for an opportunistic networking system (Store-and-Forward methodology). We can read this as how a unique message travels between nodes. Traversal is viewed from the point-of-view of node 0.

---
EOT