Mathematics & Computing
Dept Maths and Comp column background

Research

The key research program is the Doctor of Philosophy, usually known as the PhD. In order to qualify to undertake a PhD it is necessary to undertake prior studies which include a research project, for example, a masters, or an honours degree. The most common size of project in such programs at USQ, and in other Australian universities today is 4 credit points, or the equivalent of one semester working full-time. Because the passage of time is also often quite helpful in successfully completing a research project, it is often the practice that they are undertaken part-time or in parallel with course work or paid work at the university.

Research experience as provided in the research project of the Master of Computing, the BIT(Hons) or the PhD is also essential for anyone wishing to work in research and development. There is little point in conducting low quality research. Without adequate experience and skills, it is very easy to be misled about the originality and relevance of research and development, and the cost of discovering late in the piece that a piece of work is not original, or not relevant, can be catastrophic to a small project or even for a small company. Therefore, the ability to conduct research and development under "laboratory conditions" while still a student is an invaluable opportunity provided by these programs. The essential qualification for anyone wishing to make a career in research whether in universities, industry, or semi-government industries is the PhD. However, it is not necessarily essential to go straight from an undergraduate degree, to an honours degree or masters, and then to a PhD. In particular, the opportunity often exists to undertake a PhD part-time while employed in an institution where research is carried out, and this may be the appropriate path to follow for many students.

Selecting a Supervisor and a Research Project

Every research project must be supervised by a staff member at USQ in the Department of Mathematics and Computing. This applies to all projects, of whatever size. Projects can have more than one supervisor, and supervisor's may come from outside the Department, and from outside the University. At least one of the supervisors should be take a special interest in the field of the project.

Selecting a research project and selecting a supervisor are not separate activities. Students may identify and select a project without reference to a particular supervisor, and then seek a supervisor, or alternatively, a student may prefer to consider projects proposed by staff. Selecting a project which has been defined by a staff member has quite a few advantages, primary among which is that all members of academic staff have sufficient experience to identify projects which are feasible and sufficiently ambitious to form a good project, whereas the level of experience a typical student has had provides little confidence that a project proposal will shape up satisfactorily as a successful research project.

Topics for Honours, Masters, and other Student Projects

All projects are arranged by discussion between a student and a supervisor, possibly with assistance from the postgraduate coordinator. Students will select a project from the list provided on our postgraduate web pages and then approach the individual supervisor or the postgraduate coordinator. The supervisor is then at liberty to agree to work with the student to complete the proposed work as a 1, 2, or 4-credit point project.

Mathematical Projects

SupervisorProject Title
Dr. S.A. SuslovResonances in a subcritical channel flow
Dr Sergey A. SuslovBifurcations in convection
Dr. D.V. StruninSpinning and chaotic fronts in solid-phase combustion
Dr. D.V. StruninModelling oil spill confinement with floating barriers
Dr Sergey A. SuslovBuoyancy and shear competition in convection
Ron AddieQuantum Simulation
Prof Tony RobertsUnbiased estimation of multi-fractal dimensions of finite data sets
Prof Tony RobertsDevelop effective and complete low-dimensional models of fluid dynamics
Prof Tony RobertsHolistic discretisation of PDEs based upon modern dynamical systems theory

Statistical Projects

SupervisorProject Title
Dr. Karla Ziri-CastroTime Series Analysis in Wireless LAN Technologies
Dr Peter DunnStochastic modelling of rainfall
Dr Peter DunnImproved Confidence Intervals on SOI-based Rainfall Probability Charts
Ashley PlankMultiple Imputation
Ashley PlankPower analysis for nonparametric tests
A/Prof. Shahjahan KhanPredictive Inference
Paul FaheyGeometric based control charts
Dr Peter DunnNumerical derivatives for Tweedie estimation
Dr Peter DunnCensored Tweedie models
Ashley PlankEstimation of species richness
Dr Peter DunnRandom numbers for rainfall simulation from Tweedie distributions
A/Prof. Shahjahan KhanImproved Estimation
A/Prof. Shahjahan KhanEstimation of Mean Under Uncertain Prior Information
A/Prof. Shahjahan KhanEstimation of Parameters of Regression Model Under Uncertain Prior Information
A/Prof. Shahjahan KhanEstimation Under Alternative Test Statistics
A/Prof. Shahjahan KhanThe Multivariate Student-t Distribution
A/Prof. Shahjahan KhanOccupational Mobility
Dr Peter DunnImproving the deviance in generalized linear models

IT Projects

SupervisorProject Title
Ron AddieQuantum Simulation
Richard WatsonPlagiarism detection for computer programs
Ron AddieVirtual Software Development Laboratory
Zhongwei ZhangGraphical LAN Simulation using Java
Jiuyong LiData Mining Projects
Dr. Zhongwei ZhangNetwork Simulations Using NS2
Hua WangRole based access control management
Hua WangElectronic cash
Zhongwei ZhangNetwork Forensic Analysis Using Artificial Intelligence
Zhongwei ZhangIntelligent Agents for Network Management
Ron AddieNetwork Security Validation
Tony RobertsParallelise some computer algebra
Ron AddieReliable Network Design
Yan LiThe Auto-Reconstruction of Realistic Head Modelling of EEG
Yan LiAdaptive Speech Separation Using Neural Networks
Hua WangTicket based access management for M-services
Jiuyong LiBioinformatics Projects
Khaleel PetrusRelating Speaker Recognition Rate with the Contents and Length of Speech Utterance
Penny Baillie-deBylComputer Games Research
Yan LiSecure XML Documents With Signatures
Yan LiAn Intelligent Technique for Intrusion Detection in Computer Networks
Dr Yan LiARC PhD Scholarship Available

These are not necessarily the only projects staff may be interested in supervising. Another place to get some ideas about the areas where staff are undertaking research is the Mathematics and Computing Departmental Research Report.

Stages of a Successful Project

Here are the stages which it is recommended that any project proceeds through:

First Half

Second Half

Joint Projects

Certain projects are sufficiently ambitious that more than one student could effectively work on the project at once, or in succession. It is not easy to judge which projects are capable of large scale development and which are not, however, by default it should be assumed that any of the projects listed may be developed by teams of more than one student. Likewise, many of the projects might warrant joint supervision by more than one member of staff. In particular, Projects 8 and 10 listed above should be regarded as two aspects of the same project, and will probably be supervised by more than one member of staff.

Assessment

The assessment is based mainly on the dissertation. Usually the final seminar takes place after the dissertation has been submitted and the examiners will be present at this seminar. There will be two (or possibly three) examiners.