SECReT 2010 PhD projects
- Metal oxide semiconductor gas sensors as an electronic nose for the detection of microbial agents
- What are the factors that make communities vulnerable to, or resistant against, the emergence of radicalising settings?
- Covert taggant nanoparticle inks - discovery, process and product development, and analysis for sustainability and efficiency
- Diffusion processes of political violence: The role of information
- Engineering IT risk awareness, education and training
- Three-dimentional imaging of baggage for security applications.
- Understanding the traffic-driven epidemic spreading in scale-free networks
- Optimal search and detection of targets in an uncertain environment using unmanned aerial vehicle
- Explosive residue: Evaluation and optimisation of detection and sampling procedures
- Forecasting adversary’s scenarios: Systemic competitive red teaming
- Secure digital archive and web search using a Probably Approximately Correct architecture
- Mobilising community resilience through techno-social innovation
- Numerical modelling/empirical analysis of civil conflict
- Landmine, IED, UXO Detection using Ground Penetrating Radar from an Unmanned Aerial Vehicle
- Towards a usable and less disruptive security in the workplace
- Securing from exploits using information theoretical techniques
- Crime drop in Chile: Searching for causes and mechanisms
- Inferring user behaviour despite wireless network encryption
- The Chain of Evidence - a critical appraisal of the applicability and validity of forensic research and the usability of forensic evidence
Secure digital archive and web search using a Probably Approximately Correct architecture
7 March 2012
There is a growing interest in digital archives. A number of digital archive projects exist that use peer- to-peer (P2P) networks to provide long-term, persistent storage of digital content. Several are designed to withstand censorship attempts. They allow retrieval of known items, but the problem of indexing content and providing a secure search service, free of any controlling authority, remains unsolved. One solution might be to use the same P2P network to also host the search service.
This PhD will investigate and implement a distributed peer-to-peer (P2P) system that can be used to provide a search service for digital archives and the web. The system will be based on a Probably Approximately Correct (PAC) architecture. PAC search involves dividing up the document search index, distributing these partial indices randomly across nodes (computers), and performing queries by consulting a random subset of partial indices. This PhD will show that the random nature of PAC can increase security by making for an unpredictable adversarial target.