ISRD 01 Ship detection with polarimetric SAR dataMaritime surveillance with SAR (synthetic aperture radar) offers the advantages of all weather, day/night operation and sensitivity to man-made targets. In the past such radars have used a single polarisation channel but current research is assessing the benefits of fully polarimetric capabilities. While the polarimetric properties of backscatter from vessels is expected to be different to that of the ocean (owing to differences in the underlying scattering mechanisms), the best method of exploiting those differences is not clear.
Recent work by Canadian researchers has shown the benefits of applying statistical detection theory to this problem. This project aims to verify and analyse the Canadian results using data collected with Ingara - DSTO's in-house airborne SAR system.
Comparisons will be made with other detection techniques and performance will asessed using ROC (receiver operator characteristic) curves. Potential for improvements will be considered.
The project can be loosely broken into the following four components: 3 weeks induction into DSTO and background reading; 3 weeks preparing data and coding algorithms; 3 weeks applying the code and analysing the results; 3 weeks finalising the results, writing a report and preparing a presentation.
Electrical Engineering, Physics, Maths Working knowledge of Matlab, Good report writing and presentation skills Intelligence, Surveillance &
Reconnaissance Division DSTO Edinburgh SA
ISRD 02 Target Tracking and Data Fusion (2 students)Target tracking is the process where surveillance sensor measurements are used to answer questions such as "how many targets are there". In this project you will learn how to track multiple targets using measurements from multiple surveillance sensors. You will implement and test several tracking and data fusion algorithms using real and simulated data and produce graphical output.
Knowledge of state space estimation, Bayesian methods, time series analysis and Kalman filtering will be developed and results are required to be documented.
Researching and implementing algorithms in MATLAB. Testing using real and simulated sensor data. Producing and documenting results
Applied Mathematics, Statistics, Signal and Information Processing, Computer Systems Engineering, Electrical and Electronic Engineering
Programming skills, preferably using MATLAB Solid background in Mathematics (probability theory, linear algebra)
Intelligence, Surveillance & Reconnaissance Division DSTO Edinburgh SA
JOD 01 Balance of Investment Analysis of Counter Improvised Explosive Device Capability DevelopmentUse or refinement of existing Balance of Investment methodologies to explore the consequences of different levels of future CIED capability and to guide priorities for future technology research.
Understanding of quantitative and/or semi-quantitative operations research methods. Eliciting requirements to enable problem definition. Tailoring or extending methods to address problem definition. Collection of data to populate models.
Execution of models and analysis of output to provide advice for decision-making. Scientific report writing. Operations Research, Applied Mathematics
Joint Operations Division DSTO Edinburgh SA
JOD 02 Systems Analysis of Counter Improvised Explosive Device Concept DevelopmentUse of systems analysis or systems engineering methodologies to enable understanding of CIED as a system to determine CIED requirements and develop new CIED concepts.
Understanding of systems analysis or systems engineering research methods. Application of various techniques, such as process mapping, functional flow diagrams, value chains, N2 charts, influence diagrams, other architecture frameworks, to derive system conceptualizations of CIED. Analysis of products to provide advice for decision-making. Scientific report writing.
Systems Analysis, Systems Engineering Joint Operations Division DSTO Edinburgh SA
LOD 01 Using Attractors for obtaining better imagery for use in Human Identification SystemsAttractors are devices that can be used to attract a person's attention towards a target. They can be visual (i.e. a programmable message sign) or auditory (i.e. a beep). While attractors have most commonly been used in marketing and emergency signage, we have recently begun investigating their utility for attracting a person's attention towards a camera source (such as CCTV) for identification purposes. At a trial conducted earlier this year, Land Operation Division's National Security System Analysis Team collected data using both auditory and visual attractors. We intend to analyse the data to determine whether attractors are appropriate for use as aids for human identification systems (such as facial or iris recognition systems). You will become a member of a small team of research psychologists who have been leading this research.
You will conduct an indepth literature review in the area of attractors (which will span the areas of cognitive, perceptual, forensic and organisational and human factors psychology). You will also be involved in the analysis of data obtained during the trial (which includes recorded video footage and some qualitative data) In addition, you will have the opportunity to publish the research and present the findings at a DSTO seminar.
Psychology, Human Factors Well developed research, writing and analysis skills, including some basic statistical analysis experience (SPSS) would be desirable.
The National Security System Analysis team is multidisciplinary and includes engineers, mathematicians as well as research psychologists. We would ideally like to hear from people who will be entering the final year of their undergraduate degree (majoring in psychology) or Honours/postgraduate students in the same discipline.
Land Operations Division DSTO Edinburgh SA