ANU

PhD Study

about PhD study

You end up looking like this (and the males end up looking even worse):

nosferatu

current PhD students

Each corresponding to well punctured voodoo dolls (only kidding):

  1. Qiu, Lei (2008 writing up) "Wireless Localization and Spatial Resolution" (8)
    What are the fundamental limits to being able to locate a wireless mobile terminal in space given measurements over a limited region?  [Wireless Communications and Signal Processing]
  2. Zhang, Wen (2006-current) "Head Related Transfer Function Modeling" (7)
    Finding efficient functional representations to head related transfer functions for human hearing including from incomplete measurements on the sphere.  [Audio Signal Processing]
  3. Shams, Ramtin (2007-current) "Automatic Registration Methods for Multi-modal Medical Image Analysis" (10)(11)
    Multi-modal volumetric image registration using mutual information and GPU accelerated hardware processing.  [Medical Image Processing]

    Description: Registration is a problem frequently encountered in medical image analysis. It is often required to align images acquired by different imaging devices and techniques (accounting for multi-modalities) and at different times (including long periods resulting in multi-temporal images). Registration allows for comparison of complimentary features, natural or pathological differences, data fusion and compiling atlases and is used in diagnosis, treatment planning and image guided surgery to name a few applications. For certain applications it is impractical or impossible to use fiducial markers or landmarks or simply the images may have been acquired without such planning. The group of automatic registration methods align images by maximizing a similarity measure using image information over a possible range of transformation parameters.

    Since it is virtually impossible to try all combinations of transformation parameters, an optimization method such as Powell, simplex, gradient descent, etc., has to be used. There are two major limitations with automatic registration methods: (a) the robustness of the optimization is dependent on the shape of the cost function and can often lead to mis-registration due to presence of local minima, (b) the calculation of the cost function involves image transformation and calculation of the similarity measure which can be time consuming.

    We are currently looking at the application of registration for endoscopic and laproscopic surgery. A probe, equipped with an electromagnetic tracker and a ultrasound (US) transducer is navigated through the patients body and produces real-time US images. To allow a more accurate treatment of the patient we aim to register the 2D US images against a high resolution 3D CT image which has been acquired prior to the operation.

    Research Outcomes: The research aims to find methods that overcome the above limitations and allow for accurate and efficient automatic registration of multi-modal images, including MR, PET, and CT scans. To this end, we propose to estimate transformation parameters and optimize a mutual information (MI)-based cost function using the estimated parameters as the starting point. The method is expected to improves the robustness of the registration task by avoiding local minima and improve efficiency by reducing the number of iterations required for convergence of the optimization algorithm.

    Further work is aimed at real-time registration of 2D US images with pre-operative CT. The problem of registering a 2D US to a 3D volume is extremely ill-posed due to low resolution and image artifacts in US and small size of US data sample compared to the CT volume. To alleviate this problem, we propose to use a tracker to get an initial position of the ultrasound and limit the search space. We also propose to use multiple temporal frames of the ultrasound to further constrain the problem with the 3D space of the CT volume.

  4. Kodituwakku, Sandun (2007-current) "Heart Rate Variability Modelling" (7)
    Signal decompositions of heart rate data using novel time-frequency distribution kernels.[Medical Signal Processing]
  5. Zhou, Xianggyun (Sean) (2008-current) "Capacity of Time-varying Channels with Feedback" (11)
    Research aligned with an ARC Discovery Project on capacity of time-varying channels with feedback.  [Wireless Communications and Information Theory]
  6. Samarasinghe, Prasad (2008-current) "Dynamic Array Processing" (7)
    Advanced signal processing for dynamic arrays, arrays which move and sample non-uniformly in space-time.  [Sensor Array Signal Processing]
  7. Samarasinghe, Pradeepa (2008-current) "Registration and Segmentation of Medical Images"
    Efficient and robust medical image registration from multiple modalities and segmentation using dynamical models.  [Medical Image Processing]
  8. Wei, Liying (2008-current) "Signal Concentration in Spatial Domains" (12)
    Fundamental concentration of signals in space and spatial frequency for given closed spatial domains.  [Signal Processing]

completed PhD students

My experiences in PhD supervision.

  1. Pulford, Graham (1992) "Developments in Nonlinear Equalization" (1)
  2. Yamazaki, Ken (1994) "Convex Methods for Blind Equalization in QAM Transmission Systems"
  3. Dogancay, Kutluyil (1994) "Blind Error Detection and Channel Equalisation in Communication Systems"
  4. Ward, Darren B. (1996) "Theory and Application of Broadband Frequency Invariant Beamforming" (2)
  5. McGinty, Nigel C. (1998) "Reduced Complexity Equalization for Data Communication"
  6. Hasnie, Shazia (1999) "Modified Decision Feedback Equalization Techniques for Data Communications"
  7. Cheung, Peter K.P. (1999) "Blind Linear Multiuser Detection for DS-CDMA Wireless Networks" (3)
  8. Borah, Deva K. (1999) "Detection and Estimation in Digital Wireless Communications" (3)(4)
  9. Abhayapala, Thushara D. (1999) "Modal Analysis and Synthesis of Broadband Nearfield Beamforming Arrays" (2)
  10. Wang, Qi (2000) "Near Optimal Decoding for Trellis Coded Modulation" (5)
  11. Radlovic, Biljana D. (2001) "Dereverberation of Speech Signals in Acoustic Environments" (2)
  12. Jones, Haley M. (2001) "On Multipath Spatial Diversity in Wireless Multiuser Communications" (3)(4)(7)
  13. Teal, Paul (2001) "Real Time Characterisation of the Mobile Multipath Channel" (2)(6)
  14. Pollock, Tony (2003) "On Limits of Multi-Antenna Wireless Communications in Spatially Selective Channels" (7)
  15. Lim, Wee Gin (2003) "Switch-Mode Blind Equalizers for Single-User Systems" (7)
  16. Ho, Jaunty Tsun Yue (2004) "Space-Time Wireless Communications" (7)
  17. Zhang, Jian (2004) "Short-Range High-Speed Ultra Wideband Communications" (7)
  18. Trajkovic, Vladimir (2005) "Advanced Multiuser Receiver Structures" (3)
  19. Miniutti, Dino (2006) "Timing and Carrier Recovery in Frequency Selective Channels"
  20. Williams, Michael (2007) "Performance limits of Array Signal Processing" (7)
  21. Lamahewa, Tharaka (2007) "Space-Time Coding for Distributed Spatial Systems" (7)
  22. Krusevac, Zarko (2007) "Model based approach to time-varying communication channels" (3)
  23. Krusevac, Snezana (2007) "Mutual Coupling and Spatial Correlation in MIMO Communications" (3)
  24. Dickins, Glenn (2008) "Applications of Continuous Spatial Models in Multiple Antenna Signal Processing" (8)

associated PhD students

These are the students I formally co-supervised but produced no technical work output with (at least during the course of the thesis).

  1. Jana, Rittwik (1999) "Performance Evaluation of Single User and Multiuser CDMA Systems"
  2. Betlehem, Terence (2005) "Acoustic Signal Processing Algorithms for Reverberant Environments"
  3. Somaraju, Ram (2007 submitted) "Essential Dimension and Degrees of Freedom for Spatial Waveform Channels"
  4. Luo, Lin (2006-current) "OFDM"

research Masters students

Research Masters students

  1. Muthu, Suneetha (1997) "Performance Study of Qualcomm CDMA System on Satellite Mobile Channels" (5)
  2. Brooks, Michael (1998) "Pitch Modification Techniques for Sampled Voice" (2)
  3. Qiu, Lei (2001) "Improving the Performance of Concatenated Systems"
  4. Dickins, Glenn (2003) "Soundfield Representation, Recontruction and Perception"

key

Co-supervision with Brian Anderson(1), Bob Williamson(2), Predrag Rapajic(3), Brian Hart(4), Lei Wei(5), Rodney Vaughan(6), Thushara Abhayapala(7), Leif Hanlen(8), Tony Pollock(9), Richard Hartley(10), Parastoo Sadeghi(11), Jonathan Manton(12).