Funded PhD opportunities in interdisciplinary research on sensors and sensor systems
Friday 1st June, 2012
The College of Science and Engineering at the University of Glasgow is investing in interdisciplinary research on sensors and sensor systems. As part part of this initiative PhD opportunities are now available. Project 1 will be co-supervised by academics from the Schools of Mathematics and Statistics, and Geographical and Earth Sciences. Project 2 will be co-supervised by academics from Scottish Universities Environmental Research Centre (SUERC), the Schools of Physics and Astronomy and Mathematics and Statistics.
Studentship 1: Making sense of the environment
Supervisors: Marian Scott, Adrian Bowman, Claire Miller, Susan Waldron
It is uncommon that sensor networks have been designed from statistical first principles, more commonly they are designed accounting for constraints on the system such as energy constraints or infrastructure. In many situations, the location of sensors can be described as preferential, and thus not random nor necessarily representative, presenting challenges in the inferential models subsequently used. Flexible additive models over space and time are extremely useful and a growing research area in terms of handling massive spatio-temporal data sets, since they can be developed to examine the temporal and spatial trends and seasonality, to incorporate important covariates and to incorporate space/time interactions and an appropriate covariance structure for the environmental context. A recent project (Scott, Bowman and Rushworth) developed novel modelling tools based on P-splines that were computationally very efficient for a complex river network system. Further work is required to develop such models, and to a) visualise the model results, b) quantify the sources of uncertainty and c) enhance the connections to others layers of information derived from sources such as land-use.
Studentship 1 will develop the core statistical methodology and visualisation tools that lie at the centre of understanding many sensor systems- and thus will be immediately transferable being key to environmental data transformations. One output of the studentship will be a series of R packages to ensure that the knowledge can be exchanged and widely used for effective communication of results.
Studentship 2: Sensing the radiometric environment
Supervisors: David Sanderson, Marian Scott, Bjürn Seitz
The Fukushima nuclear accident and subsequent radioactive contamination in Japan have drawn attention to the need for rapid mobile sensors capable of detailed temporal and spatial mapping of radioactivity in urban and rural areas. Such systems, and their associated methodologies, are urgently required to address practical problems associated with dose assessment, targeting and validating remediation work, and providing information to build greater understanding of radionuclide dispersion and behaviour in complex environmental systems. Elsewhere the spatial, temporal and compositional characteristics of radiation exposure are also of interest to the health sector, to nuclear operators and regulators, to emergency responders, to security, defence, manufacturing and building sectors, to public bodies and to the wider scientific and public communities. SUERC and its University partners in Glasgow have been at the forefront of development and application of airborne, vehicular and ground based radiometric systems for more than 25 years. Recent developments include compact systems capable of individual deployment, or utilisation within mobile sensor networks. These have been evaluated in the UK and Japan and will be further developed within this project.
Studentship 2 will develop new sensor technologies with such compact systems, and also investigate new and sophisticated modelling strategies to analyse and represent information about the urban environment. It will combine work on existing spectrometers and data sets, with the development and appraisal of new, portable, radiation detectors and modes of deployment. Improved detector designs, incorporating expertise from environmental radioactivity studies (Sanderson), fundamental nuclear physics research (Seitz) and statistical modelling (Scott), will lead to a technology which can be widely deployed.
Applicants should identify which project they wish to be considered for. Candidates for either project should have a strong background and excellent degree in an appropriate mathematical, statistical or physical science area, and an interest in sensor development and applications. Knowledge of spatial statistics, and strong computational and programming skills would be advantageous, and essential for project 1. An understanding of environmental radioactivity and nuclear radiation detectors would be beneficial for project 2. Excellent written and oral communication skills, and strong time-management skills are also desirable.
The stipend is based on the UK Research Council rates and the studentship will be of 3.5 years duration. The start date will normally be 1st October 2012. For further information, for project 1 please contact Professor Marian Scott and for project 2, please contact Professor David Sanderson.