Find a SECReT supervisor
Information for overseas students
View SECReT animation
Download SECReT brochure

2011 PhD projects viewer

Understanding and preventing criminal disruption of infrastructure networks, focusing on railway disruption

Publication date:

Matthew Ashby

The railway network is an important public resource and part of the national infrastructure. Although very safe for passengers, every year in the United Kingdom around 270 pedestrians are struck and killed by trains. Decisions about where to deploy fatality-prevention measures are currently based on the locations of past incidents.

Secure and Robust Digital Archive Over Peer to Peer Networks

Publication date:

William Mayor

Can peer-to-peer protocols be used to form scalable, reliable and efficient digital archive systems? Worldwide digital data production rates are said to have surpassed 61 terabytes a second in 2011. With such a vast and accelerating output rate, digital archiving becomes a growing concern. Some estimates say that current storage capacity limits mean losing more than 35% of this data. Further to this, any archive solution needs to deal with issues of availability, reliability, efficiency and authorship. I propose to look at the application of peer-to-peer protocols in solving these issues. I hope to determine to what extent peer-to-peer, distributed storage can provide a secure and reliable archiving facility. By looking towards a shared model of data storage I hope to mitigate technological issues of storage medium upgrades, centralised storage weaknesses, and excessive, focused bandwidth requirements. However, introducing this model potentially creates new security concerns including: user participation, unauthorised data access or modification, and reduced reliability. Care needs to be taken to ensure that any future digital archive solution is fit for purpose. Digital archive requirements are likely to be distinct from current requirements of personal data backup or even relatively short-term archiving of business records or government data. A true solution to the problems of digital archive will ensure the continued access to the world's data, one, one hundred or one thousand years into the future.

Mathematical modelling to establish the effectiveness of countermeasures to radicalisation

Publication date:

Rosemary Penny

In the six years since the 7/7 attacks in London by British suicide bombers, the field of counter-terrorism has expanded in both academic and Government circles. Despite this much of the research  has been unscientific, with relatively little research being done into whether the countermeasures to radicalisation suggested by Government policy-makers actually work. This project seeks to use computer simulation and network modelling to establish whether measures implemented by Government departments are effective at hindering the radicalisation process, and thus whether they have had the intended impact in terms of preventing terrorism in the UK.

Is HPLC a useful addition to current Geo-Forensic Analytical Techniques?

Publication date:

Georgia McCulloch

Soil is a potentially highly valuable form of trace physical evidence, which can be used by investigators to reconstruct crimes.  HPLC has shown promising results in distinguishing forensically relevant soil samples taken from locations that are close in proximity, using small sample quantities and affordable, standard forensic analytical techniques.  This offers a distinct advantage over many geo-forensic analyses reported in the forensic literature, which have been costly, impractical or lacked ecological validity.  HPLC, which analyses the organic content of soil, can potentially be used to complement existing techniques that test the sample’s inorganic or physical characteristics, thereby increasing the evidential value of the sample.  This project aims to establish the conditions under which HPLC can be used to reliably analyse geo-forensic samples and whether it can be used in conjunction with traditional geological techniques.  The project will explore the temporal and wider spatial variability of soil chromatographic profiles by analysing samples taken from a variety of locations over extended periods of time.  The transfer and persistence of the organic components of soil will be studied by extracting samples from a variety of matrices after exposure to a range of environmental conditions.  Characteristic marker compounds observed during these analyses will also be isolated and identified. Traditional geological analyses will be performed to verify that HPLC is capable of corroborating these results, that HPLC sample preparation is compatible with other techniques and that HPLC offers added value to currently available techniques.  The project will also involve a placement at the James Hutton Institute, which has a specialist soil forensics research group that supports both criminal and civil forensic investigations.

Time-of-Flight X-Ray Compton Scatter Imaging for Cargo Security

Publication date:

Nick Calvert

Transporting goods via shipping container is the most efficient mode of transport and makes up over 90% of the world's trade. Securing these containers is therefore of high importance. Current cargo imaging techniques consist of high energy transmission imaging and low energy Compton scatter imaging. These provide two dimensional images which can be large and contain many complex, overlapping objects. In partnership with Rapiscan Systems Ltd, a new three dimensional x-ray imaging technique for cargo screening is explored by building on the work done during the MRes project.  The suggested technique combines Time-of-Flight information with Compton scatter imaging to recover depth of interaction information and hence three dimensional images. Three dimensional image formation will lead to easier image analysis, increased efficiency, and reduced false positive rates. The present work focuses on implementing the technique in both a laboratory setting and a real life setting. Using laboratory x-ray sources allows the limits of the technique to be explored in a controllable way, whilst using 'off the shelf' x-ray sources allows the technique to be used in a real life setting. It is hoped that the present work will lead to the next generation of cargo imaging systems.

Illicit Firearm Use and the Role of Firearm Procurement and Transfer Networks in England and Wales

Publication date:

Kate Gibson

Firearms criminality is different to other forms of criminality in a number of respects. Unlike many offences which can be analysed using Routine activity theory (Cohen and Felson 1979) whereby targets or victims can be protected by a ‘suitable guardian’, firearms offences frequently occur in public or even crowded areas, whereby a number of guardians exist, but remain powerless to stop the offence occurring. Furthermore, these offences are likely to be subject to more planning than many offences, with weapons scarce and therefore considerable effort made in many cases for weapon procurement. Nevertheless, research has shown firearms to be largely territorial (Gibson 2012, Braga and Kennedy, 2001), therefore it may be possible that they bear hidden resemblances to other offence categories such as residential burglary, and existing theories such as crime pattern theory (Brantingham and Brantingham).  The aim of this research is to use data from the National Ballistics Intelligence Service (NABIS) combined with Police data to explore the movement and reuse of firearms, and how this may be linked to factors associated with the offender and the offence. Of particular interest is the presence of a firearms ‘middle market’, whereby individuals lease or loan firearms to offenders for use in criminal activity. This is particularly pertinent as a Government consultation has recently ended for the creation of a new offence of ‘possession of firearms with intent to supply’ (Home Office, 2012).

Increasing Efficiency of Security Procedures to Detect Explosives on Metro Rail Networks through Analysis of Human Errors

Publication date:

Kartikeya Triapathi

The researcher is working in collaboration with RATP Veolia, and Reliance India Limited who will operate the first Metro Rail line in Mumbai, India. The research project will focus on modelling human errors that creep into security procedures in place at metro rail networks, often due to time, and revenue considerations. Explosives pose a major threat to metro rail projects across the world, and by modelling errors that undermine the efficacy of security procedures to detect them, the research will contribute in making such infrastructure systems safer against terrorist threats. Lessons will be drawn from medical science, and aviation industry – both of which have a rich research background – in modelling errors in complex procedures. Transferable solutions from these domains will contribute to understanding, and mitigating fallibility in detecting explosives on rail networks that are used by millions of passengers each year.

Nanomaterials for Security Applications

Publication date:

William Peveler

Detection of CBRNE (chemical, biological, radiological, nuclear and explosive) substances with great sensitivity and selectivity is a crucial tool for military and civil security personnel. In particular the detection of explosives is a key priority for the military and police. Explosives come in many varieties, and any detection devices must not only be able to indicate the presence of tiny amounts of material, quickly and accurately, but also the nature of the threat – what types of material are present. In addition to threat detection, the use of explosives on testing ranges presents an environmental problem, with residues leaching into groundwater. Therefore robust, portable and effective environmental monitoring must be implemented at such locations. Nanomaterials have been used to great effect in many fields, including medicine and security. Their novel properties have allowed exquisite detection of complex molecules, with selectivity and sensitivity. This study hopes to address the detection challenges described above by combining the fluorescent properties of quantum dots and other nanomaterials, with supramolecular moieties for solution based and hopefully vapour based detection of a range of molecules. We will investigate macrocycles and other supramolecular sytems for their specificity to certain CBRNE materials and environmental pollutants. The supramolecular moieties will then be linked to the nanomaterials to create a sensing system, based around turn-off or turn-on fluorescence. The project will cover aspects of organic and inorganic chemical synthesis, analytical spectroscopy and spectrometry, and systems engineering. Contact has been initiated with Dstl with the aim of obtaining access to CBRNE materials and project advice.

e-Voting security and acceptance

Publication date:

Pyrros Chaidos

The aim of this project is to improve the security and acceptance of internet and in-booth electronic voting. This includes identifying, defining and achieving such security properties, but also (with equal importance) making the appropriate decisions with regards to trade-offs between potentially incompatible security properties. Initial research in the area of e-voting security was focused on developing security notions that paralleled those of physical voting systems (such as correctness, ballot secrecy and universal verifiability to expand on auditing).  However, as new issues such as the unsupervised environment problem for internet voting were identified, newer security properties were required, namely coercion resistance. Another desirable property of systems utilising public boards of encrypted votes for transparency is that of everlasting privacy, i.e. votes will remain secret in the face of cryptographic advances.  Many of the above properties are believed (and some shown) to be incompatible which makes the evaluation of trade-offs even more important. A second goal of this project is to develop reusable cryptographical techniques that may be transferable to other domains. Much of the analysis of internet voting security falls into the domain of computer security and cryptography, and it is also likely that new models or cryptographic primitives will need to be developed to encompass potentially new functionality or that proof of concept implementations might be required. Evaluating the impact of security issues on the other hand, is not something that can be tackled using computer science alone, but requires a broader view of the voting process. This includes understanding the goals that societies set for the electoral process, as well as the behaviour of voters and other actors interested in influencing an election. Remote voting via post or other means is already the most popular form of voting in a number of regions and, given this potential for wide use, the security of internet voting will be critical factor in the democratic process.

Agent-Based Modelling of Wildlife Poaching

Publication date:

Using smartphone applications to record real-time, spatially located information from large groups of people about their perceptions of safety (fear of crime) in the built environment (London)

Publication date:

Reka Solymosi

This project will aim to measure Fear of Crime in parts of London by using citizen science/ crowd sourcing mobile applications to record real time, spatially located data from the population relating to their perceptions of safety in the environment, and identify various correlates of fear of crime. Methodologically this way of recording data builds on developments in fear of crime research, which currently encourages addressing frequency and intensity of ‘fear’ (worry?) in the population. It also adds the ability to record spatially and temporally relevant data without worrying about issues with participant recall. Sensor data (e.g. GSR, light levels, noise levels, etc.) can also be used to supplement the qualitative perception data and gather further information about fear of crime and it’s prevalence and correlates   The topic of ‘fear of crime’ is relevant as a security problem, as it has a great influence on policy decisions and interventions, and is also an accessibility issue, leading to spatial avoidance and the social isolation of “bad areas”. Mapping perceptions of safety can spatially locate areas associated with high levels of fear of crime, and identify these hotspots in order to target interventions, identify elements of the built environment (e.g.: graffiti, broken windows, narrow streets) associated with high fear of crime levels (to assess their role and recommend change). Finally it can also identify other conditions associated with fear of crime (current pedestrian density of street, purpose of journey, time of day, weather, etc.) and make for an interesting comparison study with the existing crime maps of local areas. Other outcomes of this project include the study of how and why people participate in this (crowd sourcing, swarming, free rider problem), and the feasibility of using apps for social science research.

Search UCL News