Taste of Research Summer Scholarships

2017/2018 Projects - School of Computer Science and Engineering

Computer Science & Engineering Research Areas

Related Projects

 

Computer Science & Engineering Projects

 

Algorithms


Project Title: Parameters and parameterized algorithms for disrupting covert networks in practice
Name of Supervisor: Serge Gaspers
Email of Supervisor: sergeg@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Kamran Najeebullah
Email of Joint/Co-Supervisor: k.najeebullah@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Algorithms
Applicable to other Engineering
schools/disciplines:
Abstract: The project is an empirical study of covert networks and an empirical study of algorithms for the MinIGL problem.

The first goal is to consider real-world covert networks, and compute some graph parameters for these networks, such as neighborhood diversity, treewidth, etc.

In the MinIGL problem, the input is a graph G and an integer k, and we would like to remove k vertices from G so as to minimize the inverse geodesic length (IGL) of the resulting graph. The inverse geodesic length of a graph is the sum, taken over all vertex pairs, of the inverse distance of the vertices in a vertex pair. The second goal is to implement recent algorithms for this problem, in order to see how their running times correlate with the parameters computed earlier in practice.
Research Environment: The student will collaborate with members of the Algorithms group at UNSW, Building K17.
Novelty and Contribution: This will be the first implementation of several very new parameterized algorithms for MinIGL. Comparative analysis of the experimental results of the implemented algorithms on real-world data.
Expected Outcomes: Working code of the implementation of the algorithms.
Experimental results.
Technical report and interpretation of results.
Reference Material Links: Haris Aziz, Serge Gaspers, and Kamran Najeebullah. Weakening Covert Networks by Minimizing Inverse Geodesic Length. Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI 2017). http://www.cse.unsw.edu.au/~haziz/minigl.pdf
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: The effect of random choices on algorithmic running times in practice
Name of Supervisor: Serge Gaspers
Email of Supervisor: sergeg@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Edward Lee
Email of Joint/Co-Supervisor: e.lee@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Algorithms
Applicable to other Engineering
schools/disciplines:
Abstract: Recently, the theoretical worst-case running times of many exponential-time algorithms have been improved by combining randomized algorithms with deterministic Fixed Parameter Tractable (FPT) subroutines.

In this Taste of Research summer project the student will implement and compare different random selection methods such as (1) no random selection, (2) uniform, (3) biased; and experimentally determine best threshold to switch from a random selection to a deterministic FPT algorithm in an empirical manner.

The student would also be required to either source, or generate sample problem input instances for the implemented algorithms, and scientifically justify their choices.

We will focus on the Feedback Vertex Set problem and the Vertex Cover problem, where we'd like to find k vertices in a graph whose deletion makes the graph acyclic or edge-less, respectively.

Research Environment: The student will work with researchers from the Algorithms group in the K17 building at UNSW.
Novelty and Contribution: This project will design and develop and experimental method and approach investigating work for which theoretical results have been published. This will give insight into practical issues which may not have been brought to light in the theoretical context. The implementation will act as an empirical study of the real-world efficiency of theoretical algorithms for NP-hard problems.
Expected Outcomes: At the end of the project the student is expected to have completed the following:
- A software implementation of a random selection framework, with determinisitic FPT subroutines. The student will also be responsible for generating input instances for the algorithm.
- Empirical experimentation of theoretical work using the software implementation. This will involve finding threshold values at which it is most useful to change from a random selection protocol to an FPT algorithm.
- A technical report outlining the considerations and design choices made and a statistical summary of the results from the experimentation outcome.

Reference Material Links: Fedor V. Fomin, Serge Gaspers, Daniel Lokshtanov, and Saket Saurabh. Exact Algorithms via Monotone Local Search. Proceedings of the 48th ACM Symposium on Theory of Computing (STOC 2016), ACM, pages 764-775. https://arxiv.org/abs/1512.01621

Serge Gaspers and Edward J. Lee. Exact Algorithms via Multivariate Subroutines. Proceedings of the 44th International Colloquium on Automata, Languages and Programming (ICALP 2017), Track A. https://arxiv.org/abs/1704.07982
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Artificial Intelligence


Project Title: Deep Learning on Polarised Radar Imagery
Name of Supervisor: Arcot Sowmya
Email of Supervisor: sowmya@unsw.edu.au
Name of Joint/Co-Supervisor: Matthew Gibson
Email of Joint/Co-Supervisor: matthew.gibson1@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Civil & Environmental Engineering
Electrical Engineering & Telecommunications
Abstract: Radar sensors on board satellites and aircraft are widely used to study the Earth’s surface (although they are also used for other planets too!). Several benefits of imaging radar over other types of sensors include their use at night, in bad weather, and ability to penetrate vegetation canopies. These benefits mean that imaging radar is widely used for mapping, tracking, environmental monitoring, and studies of land use and urban areas. Despite these attractive features, radar also poses certain challenges when applying computer vision to interpret these generated images. Perhaps most disruptive to tasks such as semantic segmentation is speckle noise but foreshortening, layover, and radar shadows also interfere with the computer based interpretation of this imagery.

By incorporating additional data such as the polarisation of the returned radar wave more information can be extracted from radar images. This project will investigate the feasibility of using polarised radar imagery (PoLSAR) in conjunction with convolutional neural networks to classify land use types. This has been explored for instance in [1].
Research Environment: The CSE group consists of an academic researcher, 8 PhD students working on machine learning and computer vision, and collaboration with 2 academic researchers at School of Civil and Environmental Engineering with expertise in PoLSAR imaging. The group uses two powerful GPU servers with deep learning software environments installed.
Novelty and Contribution: Combining PoLSAR processing with deep learning is novelustralian context, and the development of algorithms will progress both areas of research.
Expected Outcomes: Algorithms and deep networks to process PoLSAR data
Reference Material Links: [1] Zhou, Yu, et al. "Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks." IEEE Geoscience and Remote Sensing Letters 13.12 (2016): 1935-1939.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Evolutionary Art and Music
Name of Supervisor: Alan Blair
Email of Supervisor: blair@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Oliver Bown
Email of Joint/Co-Supervisor: o.bown@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Abstract: The aim of this project is to explore new ways of automatically generating art, images or music. Methods may include generative adversarial networks, long short term memory, and/or artist critic co-evolution. In previous work, we have used artist-critic coevolution to evolve line drawings using genetic programming (GP) for both artist and critic [1] and, more recently, using a GP for the artist and a deep convolutional neural network for the critic. This work could be extended in a number of different directions.
Research Environment: You will be working with a deep learning research team at CSE with experience in evolutionary art, in collaboration with a team in the Faculty of Art and Design with extensive experience in computer music and computational creativity.
Novelty and Contribution: The project could lead to novel research in computational creativity, coevolutionary dynamics, or adversarial training.
Expected Outcomes: The project could produce a functioning evolutionary art or music system, contribute to an art installation, or lead to a publication in a computational creativity conference.
Reference Material Links: [1] D. Vickers, J. Soderlund & A. Blair, 2017. Co-Evolving Line Drawings with Hierarchical Evolution, Australasion Conference on Artificial Life and Computational Intelligence, LNAI 10142, 39-49.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: General Game Playing
Name of Supervisor: Michael Thielscher
Email of Supervisor: mit@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Abdallah Saffidine
Email of Joint/Co-Supervisor: abdallahs@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Abstract: General Game Playing is the design of artificial intelligence programs to be able to play more than one game successfully. For many games like chess, computers are programmed to play these games using a specially designed algorithm, which cannot be transferred to another context. For example, a chess playing computer program cannot play checkers.
Research Environment: A General Game-Playing System is one that can understand a description of a game and play the game effectively without human intervention. General Game Playing expertise depends on intelligence on the part of the game player and not just intelligence on the part of the programmer of the game player. For this reason, General Game Playing is a good area in which to develop and demonstrate Artificial Intelligence techniques.
Novelty and Contribution: You will be working with a small international project team which has developed a former General Game-Playing World Champion, and you will bring in your own ideas on how to make this player even smarter. You will also get the chance to invent your own game and let existing systems from around the world compete in that game.
Expected Outcomes: Although this is a fun project, the more serious background is that General Game Playing is considered an AI Grand Challenge, and a general game-playing system, if well designed, would be able to help in other areas, such as providing general intelligence for robots.
Reference Material Links: http://en.wikipedia.org/wiki/General_Game_Playing

http://www.general-game-playing.de/

http://games.stanford.edu/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: General Problem-Solving Robot
Name of Supervisor: Michael Thielscher
Email of Supervisor: mit@cse.unsw.edu.au
Name of Joint/Co-Supervisor: David Rajaratnam
Email of Joint/Co-Supervisor: daver@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Abstract: General Problem Solving is the design of artificial intelligence systems to be able to solve a variety of different problems without having been pro-programmed for every single one of them
Research Environment: A General Problem-Solving Robot is one that can understand a description of a problem and solve it without human intervention. General problem-solving expertise depends on intelligence on the part of the robot and not just intelligence on the part of its programmer. For this reason, General Problem-Solving Robots are a good area in which to develop and demonstrate Artificial Intelligence techniques.
Novelty and Contribution: You will be working with a small international project team, and you will bring in your own ideas on how to apply a basic AI technique to a general-purpose, stationary, two-armed Baxter robot. You will also get the chance to invent your own general tasks for our Baxter to solve.
Expected Outcomes: Although this is a fun project, the more serious background is that General Problem Solving is considered an AI Grand Challenge, and a general problem-solving robot, if well designed, would be able to help in many different areas.
Reference Material Links: http://en.wikipedia.org/wiki/Baxter_(robot)
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Generative Adversarial Networks in Medical Imaging
Name of Supervisor: Arcot Sowmya
Email of Supervisor: sowmya@unsw.edu.au
Name of Joint/Co-Supervisor: Upul Senanayake
Email of Joint/Co-Supervisor: upul.senanayake@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Electrical Engineering & Telecommunications
Abstract: Deep learning has been of interest recently to Computer Vision, Machine Learning, as well as medical imaging communities. The promise to eliminate the need to hand-engineer features is a major attraction[1].

With deep learning, Generative Adversarial Networks (GAN) is a collection of two models; a generative model and a discriminative model[2]. This is analogous to a team of counterfeiters producing fake currency pitted against a team of policemen specialized in counterfeit detection. The generative model is the counterfeiter attempting to generate currency (samples) that can satisfy the discriminative model (police). The competition between the two models ultimately leads to a solution where the generative model can generate samples that are indistinguishable from the genuine samples by the discriminative model.

The GAN model can potentially provide solutions for a range of problems in medical image analysis. This project will implement a GAN model for medical images and develop at least one application around it.

Research Environment: The medical imaging group at CSE consists of one academic researcher, 8 PhD students and one postdoctoral researcher. The group has exellent collaborations with multiple medical specialties at Prince of Wales Hospital, which is adjacent to the university. The group also interacts with other machine learning researchers in the school. It has 2 powerful GPU servers with deep learning environments installed.
Novelty and Contribution: The paucity of annotated data is a major issue in medical image analysis as annotations are both expensive and time consuming. GANs can be used to generate new data points using the available annotated data so that the dataset is enriched. Researchers often have to discard part of their dataset due to lack of spatial resolution[3]. GANs can be used to improve the spatial resolution of a medical image. GANs are also demonstrated to be better at segmentation[4]. They can be adapted to perform registration of medical images as well. GANs can be modified to take native characteristics of medical images into consideration such as rotational invariance. Another advantage of GANs is that the networks can be modified better to suit the needs since two networks are trained.
Expected Outcomes: implementation of a GAN model specialised to medical imaging; one application around it.
Reference Material Links: References
[1] J. Schmidhuber, “Deep Learning in Neural Networks: An Overview,” pp. 1–88, 2014.
[2] I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative Adversarial Nets,” Adv. Neural Inf. Process. Syst. 27, pp. 2672–2680, 2014.
[3] C. Ledig, L. Theis, F. Huszár, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang, and W. Shi Twitter, “Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network.”
[4] P. Luc, C. Couprie, and L. J. Kuntzmann, “Semantic Segmentation using Adversarial Networks,” arxiv, Nov. 2016.


Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Hierarchical Evolutionary Computation
Name of Supervisor: Alan Blair
Email of Supervisor: blair@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Aleksander Ignjatovic
Email of Joint/Co-Supervisor: ignjat@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Sciences – Maths, Physics, Chemistry
Abstract: The aim of this project is to explore new enhancements and applications for an evolutionary computation paradigm known as Hierarchical Evolutionary Re-Combination Language (HERCL), which combines features from linear and stack based genetic programming. HERCL programs have previously been evolved for classification, control tasks, string processing [1] and evolutionary art.
A benchmark suite of Program Synthesis problems has recently been proposed in [2] which would provide suitable tasks for this project.
Research Environment: You will be working with an evolutionary computing and deep learning research team in CSE.
Novelty and Contribution: The project could produce novel research in evolutionary computation, coevolutionary dynamics and program synthesis.
Expected Outcomes: This research could lead to a publication in an evolutionary computation conference such as CEC or GECCO.
Reference Material Links: [1] J.Soderlund, D.Vickers & A.Blair, 2016. Parallel Hierarchical Evolution of String Library Functions, Parallel Problem Solving from Nature, LNCS 9921, 281-291.
[2] T.Helmuth & L.Spector, 2015. General Program Synthesis Benchmark Suite, Genetic and Evolutionary Computation Conference (GECCO 2016), 1039-1046.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Learning Ethical Behaviours by Demonstration in Social Robots
Name of Supervisor: Maurice Pagnucco
Email of Supervisor: morri@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Claude Sammut
Email of Joint/Co-Supervisor: claude@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Mechanical & Manufacturing Engineering
Abstract: Advances in robotic hardware and software have us on the cusp of robots inhabiting complex social settings; working in close cooperation with humans in homes and work places. However, there remain two significant obstacles to realising this goal. The first is the ability to efficiently develop robot programs in social settings. The second is to ensure that these programs result in social robot behaviours that cooperate safely with human collaborators and the environment. This project will develop techniques for learning robot programs by demonstration and subsequently ensuring that the resulting robot behaviours collaborate safely with humans in complex social environments by following stipulated ethical principles.
Research Environment: Robotics lab on Level 5 of J17 (Ainsworth) Building.
Novelty and Contribution: This project will demonstrate the learning of robot programs in a complex environment inhabited by humans and potentially other robots.
Expected Outcomes: The project will learn programs by demonstration for robots to operate in a domestic environment inhabited by humans and, possibly, other robots.
Reference Material Links: RoboCup@Home Standard Platform League: http://www.robocupathome.org/athome-spl
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: MedNet: A repository of pre-trained deep networks for medical images
Name of Supervisor: Arcot Sowmya
Email of Supervisor: sowmya@unsw.edu.au
Name of Joint/Co-Supervisor: Upul Senanayake
Email of Joint/Co-Supervisor: upul.senanayake@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Electrical Engineering & Telecommunications
Abstract: The advent of deep learning has had considerable impact on computer vision and medical image analysis[1], with its promise to extract features without having to hand-engineer them which is an arduous task[2].

Deep learning techniques, especially convolutional neural networks (CNN), are difficult to train due the large number of hyper-parameters that require tuning, the scarcity of methods to optimize tuning and the time to train from scratch. Therefore transfer learning, which allows adaptation of a network already trained on a different dataset to be tuned on a new one has become popular. Using transfer learning, networks can also be trained with relatively small amount of data.

In this project, an enhancement to transfer learning to adapt deep learning techniques for medical images will be implemented[3]. Techniques to overcome the aforementioned issues will be developed, and pre-trained networks trained with medical images will be created. It has been shown that pre-trained networks exhibit best performance when the dataset used to pre-train the network is similar to the new dataset[4]. We have access to multiple medical imaging modalities such as X-ray, magnetic resonance images (MRI), computed tomography (CT) and ultrasound (US). Subsequently, we also plan to localize these networks to specific anatomical structures of the body such as brain, lung, heart, limbs, breast and prostate.
Research Environment: The medical imaging group at CSE consists of an academic researcher, 8 PhD students and a postdoctoral fellow. The group has excellent collaborations with multiple medical specialties at Prince of Wales Hospital, which is located adjacent to the university. The group also beenfits from interactions with other machine learning researchers in the school, and runs a weekly group meeting. The group has 2 powerful GPU servers for research purposes, with necessary deep learning environments installed.
Novelty and Contribution: As pre-trained networks using medical images are not yet available, networks trained on natural image datasets are used, such as ImageNet. However, natural images are typically colour images (RGB) and represent natural scenes, while medical images are grayscale and represent human anatomical structures. Medical images also have specific characteristics such as rotational invariance, which could be exploited to further optimize the networks. In this project, an enhancement to transfer learning to adapt deep learning techniques for medical images will be implemented[3].
Expected Outcomes: Mednet, a pre-trained deep network for medical images.
Reference Material Links: References
[1] P. Mamoshina, A. Vieira, E. Putin, and A. Zhavoronkov, “Applications of Deep Learning in Biomedicine,” Mol. Pharm., vol. 13, no. 5, pp. 1445–1454, May 2016.
[2] J. Schmidhuber, “Deep Learning in Neural Networks: An Overview,” pp. 1–88, 2014.
[3] P. Baldi, “Autoencoders, Unsupervised Learning, and Deep Architectures,” ICML Unsupervised Transf. Learn., pp. 37–50, 2012.
[4] N. Tajbakhsh, J. Y. Shin, S. R. Gurudu, R. T. Hurst, C. B. Kendall, M. B. Gotway, and J. Liang, “Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?,” IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1299–1312, 2016.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Predicting Rare Events / Learning from Imbalanced Data
Name of Supervisor: Arcot Sowmya
Email of Supervisor: sowmya@cse.unsw.edu.au
Name of Joint/Co-Supervisor: David Muchlinski
Email of Joint/Co-Supervisor: d.muchlinski@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Predicting rare events, such as genocide onset or mass atrocities, is quite challenging but much desired. Only with adequate lead times and proper political mechanisms can such tragedies be avoided. This project focuses on improving on such rare event predictions by exploring machine learning from imbalanced datasets and customizing frameworks for Genocide/Mass Atrocity forecasting.
Research Environment: This research will take place under Atrocity Forecasting Project (a joint collaboration between researchers of UNSW and USyd).
Novelty and Contribution: There exists no established or standardized forecasting technique for Atrocity/Genocide prediction. This project hopes to establish grounds for such a standard.
Expected Outcomes: A working forecasting technique/framework that practically improves upon the previously obtained genocide forecasting results. A research paper and complete project report.
Reference Material Links: 1) Verdeja, Ernesto. "Predicting Genocide and Mass Atrocities." Genocide Studies and Prevention: An International Journal 9.3 (2016): 5. 2) Davison, Anthony C., and R. Huser. "Statistics of extremes." Annual Review of Statistics and its Application 2 (2015): 203-235. 3) Hastie, Trevor, Robert Tibshirani, and Martin Wainwright. "Statistical Learning with Sparsity." (2015), CRC. 4) He, Haibo, and Edwardo A. Garcia. "Learning from imbalanced data."Knowledge and Data Engineering, IEEE Transactions on 21.9 (2009): 1263-1284. 5) Semenovich, Dimitri, Arcot Sowmya, and Benjamin E. Goldsmith. "Predicting onsets of genocide with sparse additive models." ICPR. 2012. 6) Goldsmith, Benjamin E., et al. "Forecasting the onset of genocide and politicide: Annual out-of-sample forecasts on a global dataset, 1988–2003." Journal of Peace Research (2013): 0022343313484167.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: RoboCup Soccer
Name of Supervisor: Claude Sammut
Email of Supervisor: claude@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Maurice Pagnucco
Email of Joint/Co-Supervisor: morri@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Abstract: RoboCup is the world's premier international robotics competition, held in different countries each year. The competition promotes cutting edge research in robotics and artificial intelligence and aims to encourage students to take up work in these areas. UNSW has been one of the leading teams in the Standard Platform soccer league, winning the competition five times. This project will develop new techniques in computer vision, robot motion and navigation and decision making. Many of these techniques are applicable beyond robot soccer, so the student will learn skills that are valuable across a wide variety if applications.
Research Environment: UNSW-CSE has many years of experience in RoboCup. The laboratory has a full soccer field and a team of Nao humanoid robots, that are used in the standard platform competition. The UNSW team has developed a code-base that has been adopted by other teams in the competition and this all serve as a starting point for this project.
Novelty and Contribution: The scholar will be able to chose topics in a variety of areas that are all innovations in robotics and AI, these include improvements in the robot vision system, localisation, behaviours and locomotion.
Expected Outcomes: The outcomes will be working code for modules selected from the above research areas and a report, documenting that code.
Reference Material Links: See http://www.cse.unsw.edu.au/~robocup. In addition, proceedings from the RoboCup Symposium are published by Springer each year.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: RoboCup@Home - Robots in the home
Name of Supervisor: Maurice Pagnucco
Email of Supervisor: morri@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Claude Sammut
Email of Joint/Co-Supervisor: claude@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Artificial Intelligence
Applicable to other Engineering
schools/disciplines:
Mechanical & Manufacturing Engineering
Abstract: This project is to work with the RoboCup@Home team on developing new autonomous software for domestic service robots, specifically software for the Toyota Human Support Robot (HSR). The competition and research project looks at tasks that a domestic robot must be able to perform. These tasks include:
- Long-term Navigation and Mapping in office environments with people, and dynamically changing environments.
- Object recognition and classification of previously unknown objects, or identification of objects from similar previously observed objects.
- Object manipulation and object grasping with a robot arm.
- Opening and closing office or cupboard doors.
- Speech recognition and human-robot conversation, in open noisy environments.

There is a wide scope for research projects within the team, therefore students will be able to choose what they are interested in.
Research Environment: The CSE Robotics Lab is located on Level 5 of the J17 (Ainsworth) Building. There are several robot platforms available. This project will use the Toyota Human Support Robot (HSR).
Novelty and Contribution: Developing software that allows a robot to safely interact with its environment and the people that live in that environment is a significant challenge. This project will develop solutions to this problem.
Expected Outcomes: This project will focus on one aspect of developing software that allows a robot to operate in a domsetic environment and interact with people (and potentially other robots).
Reference Material Links: RoboCup@Home Standard Platform League: http://www.robocupathome.org/athome-spl
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Databases


Project Title: Efficiently Processing Structure-based Queries over Large Graph Data
Name of Supervisor: Sci Professor Xuemin Lin
Email of Supervisor: lxue@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Dr Wenjie Zhang
Email of Joint/Co-Supervisor: Wenjie.zhang@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Databases
Applicable to other Engineering
schools/disciplines:
Abstract: Recent years have witnessed rapid developments of social media, online communities, mobile communications, transportation, and collectively produced knowledge resources. Accumulated digital data often involve huge volumes and complex relationships that are modeled by graphs with vertices to represent objects and edges to represent relationships. Advances in electronic data collections are leading to an exciting new research area – Big Data. Driven by a number of key applications, this taste of research program aims to study pattern-based structure matching over large graphs. The problems involved are computationally hard (NP-Complete or NP-Hard). The investigation will study representative pattern-based structure queries by developing novel and efficient computation paradigms and algorithms. Empirical study will be conducted on both real and synthetic large graph datasets.
Research Environment: The database group in CSE consists of five faculty members, two research fellows, and about 20 PhD students. It is arguably an internationally leading research group in the field of Databases. This is evidenced by the consistent publications in the prestigious venues The database group in CSE consists of five faculty members, two research fellows, and about 20 PhD students. It is arguably an internationally leading research group in the field of Databases. This is evidenced by the consistent publications in the prestigious venues from the group every year, and the many tutorials delivered at the top-tier conferences in the fields, including KDD08, SIGMOD08, SIGMOD09, and ICDE11.
Novelty and Contribution: -Novel theorems and analysis techniques will be developed to contribute to big graph analytics.
-In addition to theoretical research, comprehensive empirical study results will be reported to show the practical performance of developed techniques on both real and synthetic data sets.
-Finally, a system prototype will be developed to show the efficiency and effectiveness of our techniques.

The student will mainly contribute to algorithms design and system prototype development.
Expected Outcomes: The success of the project will significantly contribute to the technology development and lay the scientific foundation of Big Data. Anticipated outcome includes a set of new theorems and novel data processing techniques.
Reference Material Links: W. Fan, X. Wang, and Y. Wu. Diversified top-k graph pattern matching. PVLDB, 6(13):1510–1521, 2013.
2013.

Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Embedded, Real Time & Operating Systems


Project Title: Capability-oriented programming interface for seL4
Name of Supervisor: Gernot Heiser
Email of Supervisor: gernot@unsw.edu.au
Name of Joint/Co-Supervisor: Kevin Elphinstone
Email of Joint/Co-Supervisor: k.elphinstone@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: The formally verified seL4 microkernel is arguably the world's most secure operating system. Capability-based access control is a core enabler of security, as it provides fine-grained control over access rights.

Existing middleware for seL4 provides a programming model similar to traditional (Posix-like) systems, effectively reverting to ambient authority and thus abandoning most of the security advantages of capabilities.

The recently introduced object-capability features in JavaScript, the language widely used for web pages as well as in the embedded space, provides the opportunity for a programming interface to seL4 that avoids the above sacrifice.

This project is to develop and evaluate a JavaScript embedding of the seL4 API, using the "embedded" XS JavaScript virtual machine.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: The first programming interface of a high-security OS in a modern and widely-used programming language will open the way for using seL4 for secure server-to-server web services, such as secure escrow for web transactions.
Expected Outcomes: A proof-of-concept JavaScript embedding of the seL4 API, and a performance evaluation
Reference Material Links: https://seL4.systems
https://github.com/Kinoma/kinomajs
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Fuzz test seL4 systems
Name of Supervisor: Ihor Kuz
Email of Supervisor: ihor.kuz@unsw.edu.au
Name of Joint/Co-Supervisor: Gernot Heiser
Email of Joint/Co-Supervisor: gernot.heiser@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: The seL4 kernel is formally verified, so we have high confidence that it doesn't have (many) bugs. Other parts of the seL4 platform (including libraries, drivers, etc.) have not yet been verified, so may have (possibly many) bugs in them. The goal of this project is to use a fuzzer (or any other technique) to analyse some of this non-verified code to find bugs and potential security vulnerabilities in it. If you are so inclined you can create exploits for these bugs and/or fix them too.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: This work contributes to hardening the seL4 platform and our overall project for developing trustworthy software systems.
Expected Outcomes: A report of vulnerabilities found in the non-verified part of the seL4 platform, and possibly exploits and/or fixes for those vulnerabilities.
Reference Material Links: Students should have done an Operating Systems course. Prior experience with seL4 is a bonus.

http://sel4.systems/
https://ts.data61.csiro.au/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Graphical Editor for Building Componentised Operating Systems
Name of Supervisor: Ihor Kuz
Email of Supervisor: ihor.kuz@unsw.edu.au
Name of Joint/Co-Supervisor: Kevin Elphinstone
Email of Joint/Co-Supervisor: kevin.elphinstone@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Abstract: DATA61's CAmkES (a component-based platform for developing microkernel-based systems on seL4) uses an Architecture Description Language (ADL) to describe the software architecture of an operating system. While the ADL helps to ease the difficulty of designing and building such a system, ADL documents quickly become too complicated to read and manipulate (in a text format) when the operating system becomes non-trivial. The goal of this project is to develop a graphical editor to design such componentised operating systems: allowing users to draw new components and connections and manipulate existing ones, then generate the code that represents their drawn system.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: This work will contribute to the CAmkES platform and our overall project for developing trustworthy software systems.
Expected Outcomes: A graphical editor for designing and developing component-based operating systems.
Reference Material Links: Students should have done an Operating Systems course. Prior experience with CAmkES and/or seL4 is a bonus.

https://wiki.sel4.systems/CAmkES
https://ts.data61.csiro.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Improving the software architecture of DNA analysis
Name of Supervisor: Sri Parameswaran
Email of Supervisor: sri.parameswaran@unsw.edu.au
Name of Joint/Co-Supervisor: Arash Bayat
Email of Joint/Co-Supervisor: a.bayat@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Abstract: At UNSW we have developed novel algorithm for DNA analysis which will enable multi processor systems to execute many times faster than what is presently available. This algorithm has to be translated to executable code which is optimal for small and efficient multi-processors. The candidate is expected to perform this task which is challenging due to the sheer sizes of data that has to be managed effectively.
Research Environment: State of the art tools and access to world renowned researchers.
Novelty and Contribution: The optimal creation of the new sequence alignment algorithm.
Expected Outcomes: Papers and software.
Reference Material Links: See DNA analysis papers.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Learning Hardware
Name of Supervisor: Sri Parameswaran
Email of Supervisor: sri.parameswaran@unsw.edu.au
Name of Joint/Co-Supervisor: Haseeb Bokhari
Email of Joint/Co-Supervisor: h.bokhari@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Convolutional Neural Networks are a type of learning system which can be used to identify images. In this project the candidate will implement these systems on FPGAs. There are multiple challenges. These systems are large, and we have limited hardware. We would need to reduce the hardware consumption without compromising the real time deadlines.
Research Environment: We have a large group of excellent researchers available. We also have state of the art tools and equipment available.
Novelty and Contribution: New Hardware models
Expected Outcomes: A working system on FPGAs.
Reference Material Links: See CNN related publications.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Performance limits of real-time operating systems
Name of Supervisor: Gernot Heiser
Email of Supervisor: gernot@unsw.edu.au
Name of Joint/Co-Supervisor: Anna Lyons
Email of Joint/Co-Supervisor: anna.lyons@data61.csiro.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: eChronos is a formally-verified RTOS designed for deeply-embedded systems with no memory protection and single-mode execution. Sloth is a system for a similar application domain, which takes the unusual approach of leaving all scheduling to hardware, by running everything in an interrupt context. This enables the best possible performance by minimising the amount of software on the performance-critical path, but limits the use of Sloth to processors where interrupts mode can be triggered by software.

This project is to evaluate and quantify the performance advantage (if any) of Sloth over eChronos.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Sloth is presently the world's fastest RTOS. eChronos, which has the advantage of formal verification and less dependence on hardware features, is a more traditionally-designed RTOS. This project will determine whether the performance advantage of Sloth is significant enough to justify the different (and more limiting) design. The results are eminently publishable.
Expected Outcomes: A better understanding of RTOS design tradeoffs.
Reference Material Links: Trustworthy Systems group: http://trustowrthy.systems
seL4: http://sel4.systems
eChronos: https://trustworthy.systems/echronos/
Hofer et al, RTSS'09
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: ROS native on seL4
Name of Supervisor: Ihor Kuz
Email of Supervisor: ihor.kuz@unsw.edu.au
Name of Joint/Co-Supervisor: Gernot Heiser
Email of Joint/Co-Supervisor: gernot.heiser@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: ROS (robotics operating system) is a communication middleware that is widely used for programming robots. It typically runs on a fully-fledged OS, such as Linux, using sockets for communication. This makes it readily accessible, but from the security and safety point of view is a nightmare. The purpose of this project is to produce a native ROS on the seL4 microkernel, depending on a minimal trusted computing base. It involves an assessment of the OS services required by ROS, and design, implementation and evaluation of ROS/seL4.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: A minimal ROS can enable a security and safety analysis of robotics software, dramatically increasing the trustworthiness of the robots, and opening the way for deployment in critical systems.
Expected Outcomes: An seL4-based ROS implementation that can support the high-assurance autonomous trucks developed under the DARPA HACMS program. Performance evaluation against a Linux-based implementation.
Reference Material Links: Students should have done an Operating Systems course. Prior experience with seL4 is a bonus.

http://www.ros.org/
http://sel4.systems/
https://ts.data61.csiro.au/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: seL4 Inception: Nested seL4 systems
Name of Supervisor: Ihor Kuz
Email of Supervisor: ihor.kuz@unsw.edu.au
Name of Joint/Co-Supervisor: Kevin Elphinstone
Email of Joint/Co-Supervisor: kevin.elphinstone@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: seL4 systems all start with a root task whose job is to setup and start all the other parts of the system. Existing seL4 system (CAmkES, Genode, refOS, rump, etc.) all assume that they are started from a root task, and so only one of these can run at a time. It doesn't have to be this way. The goal of this project is to enable any seL4 process to pretend to be a root task so that multiple such systems can be run at once. Better still, we could even run these systems nested in each other (rump in CAmkES in Genode in CAmkES in refOS in Genode, etc. - like the movie Inception).
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: This work contributes to the seL4 platform and our overall project for developing trustworthy software systems.
Expected Outcomes: Software that enables multiple seL4 systems to run at the same time nested and along side each other.
Reference Material Links: Students should have done an Operating Systems course. Prior experience with seL4 is a bonus.

http://sel4.systems/
https://ts.data61.csiro.au/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Virtualising the seL4 ABI
Name of Supervisor: Gernot Heiser
Email of Supervisor: gernot@unsw.edu.au
Name of Joint/Co-Supervisor: Kevin Elphinstone
Email of Joint/Co-Supervisor: k.elphinstone@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Embedded, Real Time & Operating Systems
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Capabilities are opaque object references with implied access rights. A capability system should allow transparent interposition, eg for security monitoring.

In a capability-based operating system, this should allow complete virtualisation of the system-call interface. This project is to put seL4, the world's most trustworthy operating system, to the test: Is seL4 truly virtualisable? I.e. can an arbitrary program written to run natively on seL4 run correctly in a virtualised envorinment where all system calls are intercepted by a security monitor? If so, what is the inherent cost of virtualisation?

The simplese proof of virtualisability is to construct a minimal wrapper around a user process that intercepts all capabiltiy transfers in and out of the process and replaces them by endpoint capabilitie. Those endpoints are invoked by the process as if they represented kernel objects, but instead send a message to the wrapper which then invokes the actual operation on behalf of the kernel. The virtualisation is complete if the wrapper can proxy all such invokations without any knowledge of the operation the client process' intended. Validate this by running the seL4 regression test suite as the client process.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit
Novelty and Contribution: This work either demonstrates that seL4 indeed benefits from the nice properties of classical object capabilities, or else identifies deficiencies in seL4's ABI. A good job can produce publishable results.
Expected Outcomes: A minimal emulation layer for seL4 on seL4, and a performance evaluation of this layer.
Reference Material Links: https://sel4.systems
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Formal Methods


Project Title: Improving automation in concurrent software verification
Name of Supervisor: June Andronick
Email of Supervisor: jandronick@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Carroll Morgan
Email of Joint/Co-Supervisor: carroll.morgan@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Formal Methods
Applicable to other Engineering
schools/disciplines:
Abstract: Formal verification of concurrent OS code is one of the main research grand challenges of Data61’s Trustworthy Systems group. We have done initial work in modelling and verifying a small real-time operating system, eChronos. In this work, the reasoning about interleaved execution between tasks' code and interrupt code is done using a classical concurrency reasoning method, known as Owicki-Gries, empowered by the automation of a modern interactive theorem prover, Isabelle. We have further developed a framework to reason at the implementation level, either with Owicki-Gries, or with the more compositional Rely-Guarantee.
We are now exploring the verification of the multicore version of seL4, our verified microkernel, a landmark in software verification. A few approaches are being investigated, all currently involving a level of manual work. In this project you will investigate increasing the automation of practical concurrency verification, by designing suitable rules, allowing reuse of annotations, etc.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Your work will contribute to the general feasibility and scalability of practical concurrent software verification.
Expected Outcomes: Your work will directly impact the efficiency of the framework and proofs developed for the verification of concurrent OS code.
Reference Material Links: https://ts.data61.csiro.au
http://ts.data61.csiro.au/projects/concurrency/os-concurrency.pml
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Model Checking of Network Routing Protocols
Name of Supervisor: Kai Engelhardt
Email of Supervisor: kaie@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Peter Höfner
Email of Joint/Co-Supervisor: peter.hoefner@data61.csiro.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Formal Methods
Applicable to other Engineering
schools/disciplines:
Abstract: As utilisation of wireless networks becomes increasingly multimedia in nature (i.e., consisting of data and video), issues around bandwidth availability and Quality of Service (QoS) in general become increasingly important. Current wireless mesh solutions, however, do not consistently meet those requirements. It is a common belief that the failure of existing wireless mesh network systems are to a large extent due to the limitations of current network protocols.
As an effort to improve the performance of wireless networks, we model, analyse, verify routing protocols for wireless networks. If errors or shortcomings are found we also try to fix them and report back to the developers of the protocols.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of software systems and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Model checking is a technique for automatically verifying correctness properties of finite-state systems in general and WMNs in particular: given a model of a routing protocol, a model checker like UPPAAL (http://www.uppaal.org/) can test automatically whether this model satisfies a given specification.
In order to solve such a problem, both the model and the specification are formulated in some precise mathematical formalism, such as the UPPAAL input language.
This project aims at setting up a model checking environment for UPPAAL, suitable for streamlining and simplifying the model checking of routing protocols on certain network topologies.
Expected Outcomes: The expected outcome will be a user-friendly model checking environment; the environment should should be adaptable for the specification of classes of network topologies as well routing protocol formalisations.
Reference Material Links: http://ts.data61.csiro.au/projects/concurrency/
http://www.uppaal.org/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Modelling Routing Protocols
Name of Supervisor: Carroll Morgan
Email of Supervisor: carrollm@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Rob van Glabbeek
Email of Joint/Co-Supervisor: Robert.Vanglabbeek@data61.csiro.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Formal Methods
Applicable to other Engineering
schools/disciplines:
Abstract: Wireless Mesh Networks (WMNs) are a promising technology that is currently being used in a wide range of application areas, including Public Safety, Transportation, Mining, etc. Typically, these networks do not have a central component (router), but each node in the network acts as an independent router, regardless of whether it is connected to another node or not. They allow reconfiguration around broken or blocked paths by "hopping" from node to node until the destination is reached. Unfortunately, the performance of current systems often does not live up to the expectations of end users in terms of performance and reliability, as well as ease of deployment and management.
We explore and develop adaptive network protocols and mechanisms for Wireless Mesh Networks that can overcome the major performance and reliability limitations of current systems. To support the development of these new protocols, the project also aims at new Formal Methods based techniques, which can provide powerful new tools for the design and evaluation of protocols and can provide critical assurance about protocol correctness and performance. Close collaboration with industry partners ensures the use-inspired nature of the project.
The ideal applicant should be interested in applying Formal Methods and logic-based calculi in general; previous knowledge about process algebra is appreciated, but not necessary. It is also appreciated if the applicant has successfully passed one of the following courses (or any other course on Formal Methods): COMP 3151: Foundations of Concurrency, COMP 6752: Modelling Concurrent Systems, COMP 3153: Algorithmic Verification, or COMP 4161: Advanced Topics in Software Verification.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of software systems and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Classical routing protocol specifications are usually written in plain English. Often this yields ambiguities, inaccuracies or even contradictions. The use of Formal Methods like process algebra avoids these problems and leads to a precise description of protocols. To compare and evaluate different protocols, we aim at a compendium of standard routing protocol specifications in a unified language.
Expected Outcomes: So far we have modelled one of the standard protocols using process algebra, namely AODV, as well as a draft successor protocol that is currently being discussed by the Internet Engineering Task Force (IETF).

The project's work should include the formalisation of a second standard protocol such as OSLR (http://en.wikipedia.org/wiki/Optimized_Link_State_Routing_Protocol) or HWMP (http://en.wikipedia.org/wiki/IEEE_802.11s).
After a faithful specification has been given, the work could include the verification of basic properties of the routing protocol: packet delivery for example guarantees that a packet, which is injected into a network, is finally delivered at the destination (if the destination can be reached)
Reference Material Links: http://ts.data61.csiro.au/projects/concurrency/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Open Theory Import for Isabelle/HOL
Name of Supervisor: Ramana Kumar
Email of Supervisor: ramana.kumar@data61.csiro.au
Name of Joint/Co-Supervisor: Gabriele Keller
Email of Joint/Co-Supervisor: gabriele.keller@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Formal Methods
Applicable to other Engineering
schools/disciplines:
Abstract: The goal of the OpenTheory project is to allow specifications and proofs to be shared between different theorem prover implementations of higher order logic. Currently, this proof exchange format is supported by the provers HOL Light, HOL4, and ProofPower. The aim of this project is to extend the interactive theorem prover Isabelle/HOL with import facilities for the OpenTheory format, so that Isabelle users can access and re-use the large libraries of proofs written in any of these three provers.

In fact, a basic OpenTheory import facility exists for Isabelle/HOL, which can import OpenTheory article files. What is missing is a proper link up between the OpenTheory standard library and Isabelle/HOL's native library. Aligning different versions of the same logical constant between formal libraries in an efficient manner is an open research problem to which one might apply machine learning or a custom matching algorithm. The task of this project is to design and implement a robust and efficient scheme for importing OpenTheory theories into Isabelle/HOL targeting the native Isabelle libraries correctly.

The project requires knowledge of functional programming in a language such as ML, OCaml, or Haskell. The implementation language for this project is standard ML.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Proof exchange formats are an exciting new research and engineering direction for interactive provers. The contribution of this project would be to extend the family of provers that can communicate with each other.
Expected Outcomes: An extended implementation of OpenTheory import for the prover Isabelle/HOL, that targets its native libraries effectively.
Reference Material Links: Isabelle: http://mirror.cse.unsw.edu.au/pub/isabelle/
OpenTheory: http://www.gilith.com/research/opentheory/
TS: https://ts.data61.csiro.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: What's a good routing protocol? Measurements for Comparing Routing Protocols
Name of Supervisor: Peter Höfner
Email of Supervisor: peter.hoefner@data61.csiro.au
Name of Joint/Co-Supervisor: Kai Engelhardt
Email of Joint/Co-Supervisor: kaie@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: Data61 Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Formal Methods
Applicable to other Engineering
schools/disciplines:
Abstract: Routing protocols specify how routers communicate with each other, disseminate information to select routes between any two nodes on a network, and provide the basis for sending data (packets) through a network. They find applications in all types of (communication) networks, such as metropolitan area networks (MAN), local area networks (LAN), virtual private networks (VPN) or wireless mesh networks (WMN). Due to the diversity in applications dozens of (classes of) different protocols have been developed. Often there are several protocols for the same type of networks. In the case of WMNs there are for example reactive protocols, such as AODV or DYMO, and link-state protocols, such as B.A.T.M.A.N. or OLSR.
Using protocol implementations and test-bed experiments, it has been shown that protocols behave differently on different topologies.
For example in a star topology the centre has to forward (nearly) all messages and might be too busy to react in an appropriate time to some requests.
But what does this mean for the protocols involved? Is AODV better than OLSR, if there is only few network traffic? Should B.A.T.M.A.N be used in highly-connected networks only? ...
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of software systems and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: To answer these questions systematically, network topologies have to been classified, using different metrics.
Expected Outcomes: The main part of the project should be the creation of a topology zoo and its classification. Classification metrics used should include the average connectivity of a node, the network diameter, the density of the topology, and the beta index. After the classification metrics have been chosen, sample topologies for each category should be generated. While generating topologies it should also be determined if some topologies are members of different classes.
The second part of the project should demonstrate the usefulness of the classification. To achieve this, one particular routing protocol, e.g. AODV, should be analysed. In particular, it should be determined if the routing protocol behaves differently on different classes of topologies.
Reference Material Links: http://ts.data61.csiro.au/projects/concurrency/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Hardware Design, Computer Architectures, etc


Project Title: Hardware Trojan Emulation
Name of Supervisor: Sri Parameswaran
Email of Supervisor: sri.parameswaran@unsw.edu.au
Name of Joint/Co-Supervisor: Tuo Li
Email of Joint/Co-Supervisor: tuoli@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Hardware Design, Computer Architectures, etc
Applicable to other Engineering
schools/disciplines:
Abstract: Hardware Trojan has been identified as a major challenge to safe computing platforms. Such Trojans are inserted at design time or manufacturing time to compromise the safety of or security of systems. Such Trojans have been found in both military and commercial systems. This project aims to design and build an emulation system to mimic the effects of Hardware Trojans in computing systems. Such emulation systems will provide important data for testing the security aspects of the computing system under investigation.
Research Environment: Commercial Tools as well as access to world renowned researchers.
Novelty and Contribution: This project will produce novel methods for validation of Hardware Trojan Security systems
Expected Outcomes: Papers and possible patent.
Reference Material Links: See papers on Hardware Trojans.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Memory Architecture for Multi Processor System for Genomic Computation
Name of Supervisor: Sri Parameswaran
Email of Supervisor: sri.parameswaran@unsw.edu.au
Name of Joint/Co-Supervisor: Vikki Sam
Email of Joint/Co-Supervisor: vikki.g@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Hardware Design, Computer Architectures, etc
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: An embedded will be developed to execute genomic computing which will demonstrate the feasibility of desktop DNA analysis. Thus far, large server farms are used to enable such computation. This will enable every doctor's surgery to have a desktop DNA analysis system. Further it would also make the system portable enough for in-situ monitoring of patients in remote places.

One of the challenges is to prudently organise the memory, such that the memory traffic is reduced. This project will eamine methods to reduce this traffic.
Research Environment: Commercial tools for the development of multi-processor systems
Novelty and Contribution: A new solution for genomic algorithm hardware and memory organisation.
Expected Outcomes: Papers and possible patent.
Reference Material Links: See DNA analysis methods.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Providing OS services to programmable hardware.
Name of Supervisor: Oliver Diessel
Email of Supervisor: o.diessel@unsw.edu.au
Name of Joint/Co-Supervisor: Alexander Kroh
Email of Joint/Co-Supervisor: alex.kroh@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Hardware Design, Computer Architectures, etc
Applicable to other Engineering
schools/disciplines:
Abstract: Modern tightly coupled CPU-FPGA architectures are repositioning FPGA hardware from peripheral to first-class resource. While the distance between the CPU and FPGA has decreased, the distance between hardware tasks and hosted software services remains large.

Programmable hardware typically uses a software application as a proxy for services such as file-system access. Such systems suffer from high latency, as the host application must be scheduled. The ability of the Operating System to efficiently manage resources is also limited with this method.

While some approaches aim to provide a software-like system call interface to the hardware, our approach is to provide access to these services via virtual memory-mapped interfaces.

This work will improve the integration of programmable hardware in tightly-coupled CPU-FPGA systems.
Research Environment: We are a small group of 3 PhD students and 2 Post-Docs with mixed expertise in both FPGA and OS research and engineering.
We meet weekly to discuss progress, direction, and chocolate.
Novelty and Contribution: The approach uses existing methodologies in a new context. The novelty is primarily in the application and evaluation.
Expected Outcomes: Improved understanding of the relation between hardware and software.
VHDL design of a soft IOMMU IP core for Xilinx Zynq AP SoC.
Comparison between the proposed approach and the traditional proxy system.
Reference Material Links: BORPH: http://ieeexplore.ieee.org/document/4630010/
ReconOS: http://ieeexplore.ieee.org/document/4380686/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Miscellaneous


Project Title: Data Science in Digital Marketing
Name of Supervisor: Fethi Rabhi
Email of Supervisor: f.rabhi@unsw.edu.au
Name of Joint/Co-Supervisor: Ali Behnaz
Email of Joint/Co-Supervisor: ali.behnaz@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Miscellaneous
Applicable to other Engineering
schools/disciplines:
Sciences – Maths, Physics, Chemistry
Abstract: Data is a major driver of economies these days as oil was in previous century, thus the term data economy is often coined to address the positive impact data will have on economic growth. Artificial intelligence and machine learning have enabled data scientist to extract patterns in data and generate insights. By the advent of search engines and PPC (pay-per-click) platforms, marketing has become one of the areas to experience massive disruption. Data is described as the new gold for marketers given the wealth of customer and business insights it holds.
Research Environment: You will work with UNSW FinanceIT research group. Research group, under supervision of Professor Fethi Rabhi, consists of students and researchers working in the areas of Data Science, Semantic Data Modelling, Parallel and Distributed Computing, Software and Services Engineering. You will have the opportunity to get support from different researchers along this project to improve your research skills and enhance your contribution to this project. You will use exiting libraries developed by Google data scientist in your research.
Novelty and Contribution: This is a fun project consisting of both implementation and research components. You are expected to bring in your ideas to improve the current methods used in analysing performance of digital marketing campaigns. Existing methodologies leverage machine learning/statistical learning and statistics concepts. Also, an app will be developed which enables users to utilise developed methodology and extract autonomous insights. You can as well contribute to UI and app development for this project.
Expected Outcomes: You will collaborate with other team members to
1. Improve the methodology used in analysing performance of marketing campaign in spite of a change
2. Develop UI and app for the autonomous direct response model
3. [Potential] Publishing results of your research in a scientific journal
Reference Material Links: http://www.economist.com/news/briefing/21721634-how-it-shaping-up-data-giving-rise-new-economy
https://google.github.io/CausalImpact/CausalImpact.html
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Multimedia & Visual Communication


Project Title: Medical Image Visualization
Name of Supervisor: Arcot Sowmya
Email of Supervisor: sowmya@unsw.edu.au
Name of Joint/Co-Supervisor: Gihan Samarasinghe
Email of Joint/Co-Supervisor: gihan.samarasinghe@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Multimedia & Visual Communication
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Electrical Engineering & Telecommunications
Abstract: Medical visualisation is the use of computers to create 3D images from medical imaging data sets. Much of modern medicine relies on the 3D imaging that is possible with magnetic resonance imaging scanners and computed tomography (CT) scanners, which make 3D images out of 2D slices. Quantitative analysis and visualization of medical images in a way that allows effective analysis is a great challenge.
This project will study and implement some techniques for visualization of both raw and processed medical images of different modalities (including CT and MRI) on a variety of display surfaces.
Available display devices include 3-D TV, handhelds such as iPads and a display table. Access to EPICentre and its variety of projection surfaces may also be possible.
Research Environment: This work will take place in the Medical Imaging Research group, which is collaborative between CSE and Prince of Wales Hospital, including radiologists and other specialists. Currently the group consists of a CSE academic, 8 PhD students and a number of medical specialists at Prince of Wales Hospital. A collection of CT and MRI datasets is available, along with results of analysis.
Novelty and Contribution: The way medical practitioners view images is changing rapidly. In addition, researchers at remote sites can easily share research data and analyses, thereby enhancing their ability to research, diagnose, monitor, and treat medical disorders. Providing novel visualization idioms to both these groups is a major goal of this project.
Expected Outcomes: A working medical image visualization system on one or more devices.
Reference Material Links: 1. Matthew J. McAuliffe, Francois M. Lalonde, Delia McGarry, William Gandler, Karl Csaky, Benes L. Trus, "Medical Image Processing, Analysis & Visualization in Clinical Research", CBMS, 2001, Proceedings of the 26th IEEE International Symposium on Computer-Based Medical Systems, Proceedings of the 26th IEEE International Symposium on Computer-Based Medical Systems 2001, pp. 0381, doi:10.1109/CBMS.2001.941749
2. MIPAV (Medical Image Processing, Analysis, and Visualization) application: http://mipav.cit.nih.gov/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Networks, Sensor Networks, etc


Project Title: Analysis of a Blockchain Loyalty Program Trial
Name of Supervisor: Salil Kanhere
Email of Supervisor: salil.kanhere@unsw.edu.au.au
Name of Joint/Co-Supervisor: Eric Lim
Email of Joint/Co-Supervisor: eric.lim@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Networks, Sensor Networks, etc
Applicable to other Engineering
schools/disciplines:
Abstract: The aim of this research is to undertake a real-world trial to study the efficacy and uptake of using a cryptocurrency as an alternative to loyalty points for a loyalty program. Unlike traditional loyalty points, the value of digital currency fluctuates significantly, meaning the participant has earned something which could significantly appreciate or depreciate in value on a daily basis (over the last 30 days Ether currency has dropped 25% then risen 30%). The trial will give us insights into how users manage the digital currency that they earn through the program and whether earning a digital currency drives deeper loyalty program engagement than traditional loyalty points. We will also survey the users to obtain their feedback about their experience in participating in the trial. In particular, the project will focus on analysing the data collected from the trial to draw interesting insights about the usage patterns.
Research Environment: You will work closely with researchers from the Networked Systems and Security research group.
Novelty and Contribution: This is the first-of-its-kind trial of a cryptocurrency-based loyalty program. It is expected that there will be about 500 participants in the trial. This will give us access to a significant data set.
Expected Outcomes: Detailed Report of the Analysis, Software Tools
Reference Material Links: Detailed Report of the Analysis, Software Tools
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Data After Death
Name of Supervisor: Salil Kanhere
Email of Supervisor: salil.kanhere@unsw.edu.au
Name of Joint/Co-Supervisor: Arash Shaghaghi
Email of Joint/Co-Supervisor: a.shaghaghi@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Networks, Sensor Networks, etc
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Most of us live two lives - the physical one and a virtual mirror of it online. When we die, our physical existence comes to an end, but what happens to all the data and content we have created on our online accounts (FB, Twitter, Gmail, etc.)? What happens to all that we have stored in the cloud and, unknowingly, in servers across the globe?

“digital will” is now a thing. Moreover, major corporates have privacy policies that explain how they will treat data after a user is reported dead. There has been some media attention and investigation on this topic in Europe and US over the last few years. In Australia, an increasing number of small/medium IT companies exist, and they all deal with user data. Initial investigation reveals very limited has been done in this area for Australians and no company/application or website seems to deal with this important issue.

This project is motivated by an actual gap that we have noticed. Our approach is however not to deal with the legal requirements and challenges. Initially, we aim to investigate the public awareness about this issue in Sydney among different user groups. Thereafter, we will investigate and detail a list of technical requirements and solutions for businesses and users.
Research Environment: Sufficient support will be available from researchers in the Networked Systems and Security Research Group. You will work very closely with a PhD student on this project.

Novelty and Contribution: This an important issue that has rarely, if ever, been investigated within Australia. The contribution of this work depends on how ambitious the student will be. It may range from a technical report/news raising the awareness towards this issue, providing a list of solutions in the form of a website to businesses and users, building new tools such as mobile applications that user may use to setup their digital will, and more.
Expected Outcomes: Publications in the form of a scientific article at B-level venues, Application, and website for public domain use.
Reference Material Links: http://www.businesstoday.in/moneytoday/technology/online-content-profile-data-after-death/story/24252.html

https://www.forbes.com/sites/tamlinmagee/2013/11/19/what-happens-to-your-data-after-youre-dead/#5a140ed2224f

https://www.wired.com/2013/04/alt-text-data-after-death/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Donya: Next Generation of Anti-censorship Tools over SDN
Name of Supervisor: Salil Kanhere
Email of Supervisor: salil.kanhere@unsw.edu.au
Name of Joint/Co-Supervisor: Arash Shaghaghi
Email of Joint/Co-Supervisor: a.shaghaghi@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Networks, Sensor Networks, etc
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: You’ve probably heard of Tor? Psiphon? Freegate? .. these are common tools for many users around the world to access online resources and websites that may be otherwise censored to the general public. In this project, you will investigate how the most recent anti-censorship tools could improve once imported into Software-Defined Networking platform. Software-Defined Network is the most recent networking paradigm that major companies such as Google, HP and Microsoft have started adopting it. It is expected to be deployed globally due to benefits it brings for network operators.

This is cutting-edge research that will allow you to get familiar with hot topics in networking and cybersecurity. In Australia, there is now a huge interest by ISPs and alike to adopt SDN. Come and work on a project with real impact!

The student should be hands-on and:
- Knowledgeable in Computer Networks
- Passionate to read and learn about Software-Defined Networks (SDN)
- Able to program in JAVA or C (preferably both)
Research Environment: You will work closely with researchers from the Networked Security and Systems group. You will have access to state of the art SDN facilities in CSE.
Novelty and Contribution: You are dealing with a real problem in this project and the outcome of your work will be released in public domain. This is a work that has not been done before in the literature so you are leading a brand new idea.
Expected Outcomes: Publications in the form of a scientific article. Open source tools/code.
Reference Material Links: http://dl.acm.org/citation.cfm?id=2995974
https://arxiv.org/abs/1609.04514 (extended)
https://www.opennetworking.org/sdn-resources/sdn-definition
http://www.cisco.com/c/en_au/solutions/software-defined-networking/overview.html
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Next Generation of Text Editor for Organizations
Name of Supervisor: Salil Kanhere
Email of Supervisor: salil.kanhere@unsw.edu.au
Name of Joint/Co-Supervisor: Arash Shaghaghi
Email of Joint/Co-Supervisor: a.shaghaghi@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Networks, Sensor Networks, etc
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: n this project, you will be tackling the most challenging threat in today’s cyber security world: Insider Threat.
Today’s portable devices enable organizations to support anytime, anywhere data access models. However, this comes to a risk: data may be accessed from potentially unsafe locations in the presence of untrusted parties. A topic studied as “Contextual Access Control” in the literature. In this project, you will build a futuristic editor that is smart enough to grant/disallow functions of an editor depending on the user’s context. For example, if your supervisor is not around the editor will not allow you to print/email the confidential sections of a text file you are browsing.
The decision as to whether grant permissions or not is made by the organization's network. To achieve this, we leverage Software-Defined Networks (SDNs) to model the context of access requests and retrieve a centralized view of the context and make decisions. The editor will then enforce these restrictions for the user.
Research Environment: The student will work closely with researchers in the Networked Systems and Security research group at CSE. In particular, you would work closely with a PhD student. Sufficient support will be available. There is expensive and state-of-the-art SDN equipment available at CSE that student will get access to. However, most of the work will be through simulated environment.
Novelty and Contribution: This is a new topic that is based on our group research and extends previous work into practical applications that may be potentially of interest to companies.
Expected Outcomes: Publications in the form of a scientific article at B-level venues.
Reference Material Links: http://dl.acm.org/citation.cfm?id=2995974
https://arxiv.org/abs/1609.04514 (extended)
https://www.opennetworking.org/sdn-resources/sdn-definition
http://www.cisco.com/c/en_au/solutions/software-defined-networking/overview.html
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Social Network as a Vector for Malware
Name of Supervisor: Salil Kanhere
Email of Supervisor: salil.kanhere@unsw.edu.au
Name of Joint/Co-Supervisor: Arash Shaghaghi
Email of Joint/Co-Supervisor: a.shaghaghi@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Networks, Sensor Networks, etc
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: When Alice asks Bob to connect to her Linkedin account, Linkedin sends an e-mail. Bob then clicks on accept. However, as spammers know anyone can send an e-mail claiming to be Linkedin! Moreover, clicking on a link might bring you to a webpage that delivers malware, e.g., exploiting weaknesses in Acrobat Reader. In fact, GCHQ has been reported to use LinkedIn to deliver malware to its targets. In this project, you will carry out a detailed survey on how to replicate this attack and investigate different approaches to delivering such attacks in controlled settings. Some possible goals for this research are:

1. develop technology that finds who might know who in the real world. One approach is to find e-mail addresses from people who might know each other.
2. write software that asks 2 parties to connect to each other on Linkedin, but looks credible, even though “contacts list” of both parties are not known to the malware! The previous technology will be used to find who to invite to connect to each other. Note that the goal is not only to succeed in making Alice and Bob connect (by both accepting each others claimed invites), but also to deliver malware!
3. study similar attacks for social networks besides Linkedin,
4. find other weaknesses of social networks and demonstrate how these could be used to deliver malware.
Research Environment: You will work closely with researchers from the Networked Systems and Security research group.
Novelty and Contribution: You are aiming to survey the set of vulnerabilities that may enable an attacker to deliver malware through social networks. As part of this, you are also investigating whether this issue still exists since 2013 and if so, what a user may do to protect him/herself.
Expected Outcomes: Technical report. Software tools.
Reference Material Links: http://www.independent.co.uk/news/uk/home-news/gchq-used-quantum-insert-technique-to-set-up-fake-linkedin-pages-and-spy-on-mobile-phone-giants-8931528.html
http://www.spiegel.de/international/world/ghcq-targets-engineers-with-fake-linkedin-pages-a-932821.html
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Programming Languages and Compilers


Project Title: Information flow analysis for detecting vulnerabilities in Android apps
Name of Supervisor: Dr. Yulei Sui
Email of Supervisor: ysui@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Dr. Wen Hu
Email of Joint/Co-Supervisor: wen.hu@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Programming Languages and Compilers
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Android grows extremely fast during recent years, dominating over 80%
of world smartphone market by the end of 2016. Security issues stand out. Personal information and sensitive data leaked by mobile applications have become an increasingly critical problem.

It is important that application developers understand how different application components (in particular Activity, Service, and BroadcastReceiver) impact the lifetime of the application's process.
Incorrectly using these components results in leaking sensitive data, killing important app processes, poor user experience or consuming limited system resources unnecessarily.

Traditional static value-flow analysis, which uses pointer/alias analysis for modeling program control and data dependency, is useful but not enough for tracking information flow for mobile applications. Special features and semantics in Android applications, such as message sending/receiving through the internet, callbacks for system-event handling, UI interaction, and components, are the major obstacles for traditional program analysis.
Research Environment: You will be working with a world-renowned research team which has published high-quality papers and has developed a series of program analysis techniques/tools.

The project will be conducted based on existing tools developed by our research group and the work will be in close cooperation with one or more researchers and PhD students working in this area.

The necessary guidance will be provided, and you will be given the chances to make a practical impact to solve a challenging real-world problem.
Novelty and Contribution: The student is expected to model the lifecycle of an app using static program analysis during compile time to predict and report incorrect and inefficient usage scenarios for Android apps through investigating popular Android open-source programs hosted on GitHub.

This project will enhance our existing tool to locate potential security and/or performance problems for Android applications. Particularly, the student will focus on developing a precise static life cycle modeling technique.
Expected Outcomes: You are expected to build a tool for automatically detecting interesting but critical security bugs such as information leakage for Android applications.

It offers a good opportunity for you to learn about program analysis techniques based on large-scale software systems and also to develop your knowledge and skills in mobile security.
Reference Material Links: Android Developers Page
https://developer.android.com/design/downloads/index.html
Android VM
https://source.android.com/devices/tech/dalvik/
FlowDroid
http://sseblog.ec-spride.de/tools/flowdroid/
DroidSafe
https://github.com/MIT-PAC/droidsafe-src
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Preventing code reuse attacks in C++ using pointer analysis
Name of Supervisor: Dr. Yulei Sui
Email of Supervisor: ysui@cse.unsw.edu.au
Name of Joint/Co-Supervisor: Prof. Jingling Xue
Email of Joint/Co-Supervisor: jingling@cse.unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Programming Languages and Compilers
Applicable to other Engineering
schools/disciplines:
Electrical Engineering & Telecommunications
Abstract: Software is often subject to external attacks that aim to control its behavior. The majority of the attacks rely on some form of control hijacking to redirect program execution. For instance, a buffer overflow in an application may result in a call to a sensitive system function, possibly a function that the application was never intended to use.

Although the protection techniques such as DEP and ASLR have been widely deployed for preventing control-flow attacks via code injection.
Code reuse attacks, such as return-into-libc and return-oriented programming still remain a "hard-to-defend" control flow hijacking approach by reusing the code gadgets already resident in memory instead of code injection.
Research Environment: You will be working with a world-renowned research team which has published high-quality papers and has developed a series of program analysis techniques/tools.

The project will be conducted based on existing tools developed by our research group and the work will be in close cooperation with one or more researchers and PhD students working in this area.

The necessary guidance will be provided, and you will be given the chances to make a practical impact to solve a challenging real-world problem.
Novelty and Contribution: Fine-grained enforcement of CFI, however, can introduce significant overhead. The construction of an accurate control flow graph requires the use of a precise pointer analysis. This project aims to enable precise demand-driven points-to analysis to provide strong CFI guarantee for protecting virtual call attacks in C++. Additionally, the students also encourage to investigate how to leverage the static information to reduce overhead sandboxing by eliminating redundant instrumentations if they are proven to be unnecessary.
Expected Outcomes: You are expected to build a tool for automatically enforcing control flow integrity to prevent virtual call attack in C++

It offers a good opportunity for you to learn about preventing control flow attack and program analysis techniques for large-scale software systems and also to develop your knowledge and skills in computer security.
Reference Material Links: SVF
http://unsw-corg.github.io/SVF/

http://www.cse.unsw.edu.au/~ysui/papers/cc16.pdf

http://www.cse.unsw.edu.au/~ysui/papers/issta17.pdf

Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Proof Engineering on CakeML proofs
Name of Supervisor: Ramana Kumar
Email of Supervisor: ramana.kumar@data61.csiro.au
Name of Joint/Co-Supervisor: Gabriele Keller
Email of Joint/Co-Supervisor: gabriele.keller@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: Data61 Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Programming Languages and Compilers
Applicable to other Engineering
schools/disciplines:
Abstract: CakeML is a functional programming language (similar to OCaml or Standard ML) with a formal specification (a definition in logic) and a mechanically verified (proven correct) compiler and runtime system. The CakeML proofs are a substantial formal development (over 150,000 lines of proof script) that is starting to reach the size where scalability issues become relevant. The discipline of proof engineering, which is to proof development as software engineering is to large-scale software development, can be applied; this project is to do so.

Although some of the work to be done here is simply tweaking and refactoring, there are also many opportunities for creative ideas about how to structure and manage large formal developments, or for particular approaches to generating certain definitions or proofs automatically.
Research Environment: DATA61's Trustworthy Systems group are world leaders in research and engineering for providing unprecedented security, safety, reliability and efficiency for software systems. Successes include deployment of the OKL4 microkernel in billions of devices, the first formally verified OS kernel, seL4, and a complete seL4-based high-assurance system successfully embedded in an autonomous helicopters from Boeing. You will work with a unique combination of OS and formal methods experts, producing high-impact work with real-world applicability, driven by ambition and team spirit.
Novelty and Contribution: Large scale proofs are an exciting combination of software engineering methodology with formal methods.
Expected Outcomes: Improvements to the speed or size of CakeML proofs, and, ideally, new approaches to dealing with large formal developments.
Reference Material Links: CakeML: https://cakeml.org
HOL: https://hol-theorem-prover.org
TS: https://ts.data61.csiro.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Using Linguistic Techniques to Automatically Analyses Content of Social Media to Extract Trip Purpos
Name of Supervisor: Dr. Mojtaba Maghrebi
Email of Supervisor: maghrebi@unsw.edu.au
Name of Joint/Co-Supervisor: Dr. Taha Rashidi
Email of Joint/Co-Supervisor: rashidi@unsw.edu.au
School: School of Computer Science and Engineering
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Programming Languages and Compilers
Applicable to other Engineering
schools/disciplines:
Civil & Environmental Engineering
Electrical Engineering & Telecommunications
Abstract: A growing body of literature in social science has been devoted to extracting new information from social media to assist authorities in manage crowd projects. In this research it is aimed to utilize intelligent transportation services from social media. Travel demand modelling, and analyzing and/or managing the operation of the transport network require availability of detailed information of several types of agents playing role in generation of trips use the transport network. This information includes socio-demographic and economic attributes of people and firms, and the pattern of trips generated by them. Data is generally a valuable product which exhausts a large portion of the provided financial resources for planning and operating the transport system. As a result, not necessarily all metropolitan areas can afford collecting data on a monthly or yearly basis. This has resulted in emergent of innovative approaches to temporally or /and spatially transferring data and models [1] or indirectly imputing the required data from other readily accessible data source [2]. Traditionally, data for demand modelling has been collected using two major methods called: i) revealed preference (RP) surveys and ii) stated preference (SP) surveys. As a result technology has been employed to collect household travel survey data (or even count data) in a cost effective manner
Research Environment: School of Civil Eng/School of Computer Science
Novelty and Contribution: Extracting data from contents of social media to identify the user trip purposes
Expected Outcomes: Developing a platform that can predict/extract people trip purpose.
Reference Material Links: Abbasi A; Rashidi TH; Maghrebi M; Waller ST, 2015, 'Utilising location based social media in travel survey methods: Bringing twitter data into the play', in Proceedings of the 8th ACM SIGSPATIAL International Workshop on Location-Based Social Networks, LBSN 2015 - Held in Conjunction with ACM SIGSPATIAL 2015, http://dx.doi.org/10.1145/2830657.2830660

Maghrebi M; Abbasi A; Rashidi TH; Waller ST, 2015, 'Complementing Travel Diary Surveys with Twitter Data: Application of Text Mining Techniques on Activity Location, Type and Time', in IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC, pp. 208 - 213, http://dx.doi.org/10.1109/ITSC.2015.43
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


 

Projects offered by other Engineering Schools that may be of interest are:

 

Graduate School of Biomedical Engineering

 

Project Title: Optical electrode arrays for brain machine interface
Name of Supervisor: Dr Leonardo Silvestri
Email of Supervisor: l.silvestri@unsw.edu.au
Name of Joint/Co-Supervisor: Prof François Ladouceur
Email of Joint/Co-Supervisor: f.ladouceur@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): MEMS, Micro & Nano Technologies
School Research Area: Advanced Photonics
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Computer Science & Engineering
Abstract: We have just fabricated our first optical-electrode (optrode) array for the recording of biological potentials (biopotentials) and we are performing our first in vitro measurements. Such array represents a completely new class of neural recording devices and our long-term goal is to implant it in vivo as a ‘brain-machine’ interface (BMI). The first step is to prove that it can detect biological signal, from neurons or cardiac tissue.

This novel device is based on a new class of liquid crystals – so called deformed helix ferroelectric liquid crystal – that offer extraordinary sensitivity and linear response to electrical stimulations down to the microvolt range. As opposed to the standard liquid crystal, which typically exhibits strong bi-stability, these liquid crystals can smoothly, continuously and passively transduce small electrical signals into the optical domain.

This technology, entirely developed at UNSW, is actively being researched and commercialised in the context of optical sensing networks and has found strong support from industry in the areas of ocean monitoring, mine monitoring, and oil & gas distribution monitoring. In the proposed project we apply the technology to the biomedical domain for use in implantable devices.
Research Environment: The project is a collaboration between electrical (EE&T) and biomedical engineering (GSBME) at UNSW. The sensing technology was developed by the group of Prof Ladouceur (EE&T) while its application to the biomedical field is lead by Prof Lovell.

The successful completion of the first prototype relies heavily on the Australian Nanofabrication Facilities (ANFF) at UNSW and UQ while new breeds of liquid crystals are currently being developed at the ANFF node in Wollongong.

A provisional patent on the approach described above has been filed and will complement existing patent filings of the platform sensing technology. Prof Ladouceur’s group will contribute to the device fabrication and multiplexing approaches. Lovell’s group will engage in device testing and characterisation.
Novelty and Contribution: This project consists in helping to measure electrical activity in biological tissues (brain, cardiac, retinal) with our novel optrode array. The key activities are (i) building a suitable in vitro setup that combines the optical recording with the biological stimulation; (ii) optimising the optical read out to improve the signal to noise ratio; (iii) recording the extracellular potential in various biological tissues. This novel approach is currently being patented in Australia, China, US and EU.

We expect the successful candidate to engage in one or more of the above activities.
Expected Outcomes: Ideally, the outcome of this project would see the optical recording of extracellular potentials in various biological tissues.

This successful completion would bring to the candidate a valuable experience in the fields of biomedical engineering, together with exposure to advanced photonics concepts.
Reference Material Links: The optrode design is based on the following technological breakthrough:
"Sensors at your fibre tips: a novel liquid crystal-based photonic transducer for sensing systems", Journal of Lightwave Technology July 23, 2013.

The optrode array concept is illustrated here:
L. Silvestri et al., 2016, "Computational modeling of a novel liquid crystal-based optrode", in Proceedings of SPIE - The International Society for Optical Engineering, http://dx.doi.org/10.1117/12.2242866
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: Zedelef Pty Ltd, a UNSW spin-off company, is wanting to commercialise the outcome of this project. Zedelef also provide technological support and related expertise and thus will interact actively with the successful candidate.

[Back to Top]


Project Title: Development and Use of the Swarm Simulator
Name of Supervisor: John Page
Email of Supervisor: j.page@unsw.edu.au
Name of Joint/Co-Supervisor: Faqihza Mukhlish
Email of Joint/Co-Supervisor: f.mukhlish@student.unsw.edu.au
School: School of Mechanical and Manufacturing Engineering
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Air and ground vehicles
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: We have a unique swarm simulator in the school that is capable of simulating eight separate entities, which we intend to use for this work. This is the only one of its type in the world and was designed and built in-house. Each entity is loaded on a separate computer to avoid cross contamination and allow individual modifications. Our interest is in Self Organizing Swarms which consist of a number on entities with no central control. When given a mission the action of these entities leads to an emergent intelligence that allows individuals with a low level of cognitive assets to perform complex tasks. This technology is based on the behaviour of social insects such as ants and bees but is also applied by higher level animals including humans. The main difficulty is how to establish an appropriate rule set for the individual agents which is the bulk of our current research. Once we have established the rule set it then needs to be applied to a sophisticated simulator to prove its validity. This project will investigate how best to apply the rule sets to our swarm simulator and any modifications that can be made to improve its performance.
Research Environment: The student will work in the Simulation and Virtual Engineering (SAVE) Laboratory which is located in J17/311. This is where the Swarm Simulator is located. The laboratory has all the hardware and software required to carry out the project and the student one briefed will have access to all the equipment within the laboratory. The post graduate students, who will assist in the supervision are located inJ17/311 and will thus be available to assist in any way needed.
Novelty and Contribution: Swarm technology is finding more and more applications not only in Unmanned Aerial Systems, where it originated, but also in other vehicle control systems such as driverless cars and even fixed assets such as power generation and supply systems. One of the more interesting outcomes from this research is the sophistication of the emergent behaviour which has led to an alternative approach to AI. Free access to the in-house developed swarm simulator is a particular asset to this project. If consists of a cluster of twelve computers communicating and interacting which has also generated some interesting research.
Expected Outcomes: Because the taste of research program offers us opportunities to explore the research envelope in areas we would not have time to address it can lead to significant outcomes. In the past three students were included as authors on technical publications having provided substantial input. From the student own point of view this will provide the opportunity to work in a fast developing field. The student will also gain some knowledge in low level AI and interfacing computors.
Reference Material Links: System in the loop control of multiple autonomous UAVs
P Sammons, J Page
SimTecT 2008
Self-organised UAV swarms
TC Chi, J Page, H Cheng, J Olsen
AIAC15: 15th Australian International Aerospace Congress, 672
Experimentation and Validation of Vehicle Cluster Simulator Using NetLogo
PJ Sammons, J Page
SimTect2009
University of New South Wales Aircraft Cluster Simulator
MPJ Sammons, J Page
AIAC-13 Thirteenth Australian International Aerospace Congress
A Swarm Simulator
J Page, H Cheng
Journal of Unmanned System Technology 1 (2), 58-62
A Self-Organized Swarm Simulator
JR Page, H Cheng, S Goh
Proceedings of International Conference on Intelligent Unmanned Systems 8
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Microscope Image Analysis from Vine Leaves
Name of Supervisor: Mark Whitty
Email of Supervisor: m.whitty@unsw.edu.au
Name of Joint/Co-Supervisor: Scarlett Liu
Email of Joint/Co-Supervisor: scarlett.liu@unsw.edu.au
School: School of Mechanical and Manufacturing Engineering
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Air and ground vehicles
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: Have you ever wondered how leaves are structured? Understanding this is critical for monitoring plant health and detecting disease.

Building on experience in detecting water stress in grapevine leaves, this project will investigate methods for building models of the surface structure of leaves in order to determine health status. This is focussed on image processing from imprints of leaves photographed under a microscope and identifying major components and shapes automatically. Better understanding of plant health will enable farmers to make better decisions about what nutrients and irrigation to apply.
Research Environment: This research will involve working with an one academic from the Mechatronics groups at UNSW as well as a postdoc and Ph.D students. Viticulturists from the Australian Wine Research Institute and the South Australian Research and Development Institute will provide viticulture specific knowledge and guidance.

The applicant must be confident at programming in C, C++, Python or Matlab although no experience with viticulture is required. Experience with OpenCV or other computer vision libraries would be valued but can be learnt through this project.
Novelty and Contribution: This project involves the development and application of image processing algorithms for detecting structures at the microscopic level. Manual measurements are very time consuming and commercial software packages are limited in their flexibility to detect elements with a large amount of variation. As a result, improve knowledge of plant structures will allow more accurate management decisions to be taken.
Expected Outcomes: a) Software to automatically detect and visualise plant structures from microscope images
b) A conference or journal paper quantifying the robustness of the approach compared with existing or manual methods
These outcomes may be further developed as part of an undergraduate thesis project and even form the basis for a postgraduate research project.
Reference Material Links: See the group's webpage at www.robotics.unsw.edu.au/srv for background information and feel free to contact Dr Mark Whitty (m.whitty@unsw.edu.au).

The following UNSW courses contain content related to the topics studied in this project: MATH2089 Numerical Methods and Statistics, COMP2041 Software Construction: Techniques and Tools, MTRN2500 Computing for Mechatronic Engineers, COMP3411 Artificial Intelligence, and COMP9517 Computer Vision.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: Potential exists for site visits to vineyards for data collection.

[Back to Top]


Project Title: Street View for a Vineyard: Efficient Database Management for Large Scale Image Processing
Name of Supervisor: Mark Whitty
Email of Supervisor: m.whitty@unsw.edu.au
Name of Joint/Co-Supervisor: Scarlett Liu
Email of Joint/Co-Supervisor: scarlett.liu@unsw.edu.au
School: School of Mechanical and Manufacturing Engineering
Faculty Research Area (Theme): Immersive Systems and Virtual Reality
School Research Area: Air and ground vehicles
Applicable to other Engineering
schools/disciplines:
Civil & Environmental Engineering
Computer Science & Engineering
Abstract: Vineyard yield estimation influences a variety of management decisions the farm may apply to achieve uniform yield throughout a block at various times in each season. Despite being industry best-practice, the yield estimation process currently used by major companies is complex, labour intensive and time consuming. Recent work has focussed on applying image processing techniques to speed up and improve the accuracy of yield estimation.

Yields vary substantially, even by a factor of three within a single block, so imagery is needed from a large proportion of each block to achieve the necessary accuracy. With tens of thousands of vines in each block, capturing imagery of each vine individually has been automated using robots, however managing such a large amount of imagery remains a challenging problem. Taking imagery repeatedly throughout the season is the ultimate goal, further exacerbating the image storage, transfer and retrieval process. This project will investigate methods of implementing a Google Street View style management system which will allow browsing and automated analysis of large volumes of spatially and temporally related imagery.
Research Environment: This research will involve working with one senior and two junior researchers from the Mechatronics and Geographic Information Systems groups at UNSW as well as several Ph.D students and undergraduate thesis students. Viticulturists from Australia’s largest winemaking company (Treasury Wine Estates) will provide input on specific use cases.

The applicant must be confident at programming with databases and dealing with large datasets. No experience with viticulture is required.
Novelty and Contribution: This project involves the invention of methods to manage georeferenced imagery in a similar manner to Google Street View. The focus is on the development of a database style back-end from which imagery can be retrieved and on which external image processing techniques can be applied (said image processing techniques being outside the scope of this project) and the results recorded. A front end for browsing imagery may also be investigated. Both provide opportunities for long term development and publication.
Expected Outcomes: a) Software design and prototype implementation for a georeferenced image management system.
b) A technical report reviewing the existing approaches to this problem which may form the basis of a publication.
A successful prototype system will be follow on from an externally funded research project involving vineyards all over Australia and internationally. These outcomes may be further developed as part of an undergraduate thesis project and even form the basis for a postgraduate research project.
Reference Material Links: The following UNSW courses contain content related to the topics studied in this project: COMP2041 Software Construction: Techniques and Tools, COMP3411 Artificial Intelligence, COMP3311 Database Systems, COMP3421 Computer Graphics, COMP6714 Information Retrieval and Web Search.

See the group's webpage at www.robotics.unsw.edu.au/srv for background information and feel free to contact Dr Mark Whitty (m.whitty@unsw.edu.au).
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: Potential site visits to a vineyard near Orange

[Back to Top]


Project Title: Manufacturing blood in a dish
Name of Supervisor: Robert Nordon
Email of Supervisor: r.nordon@unsw.edu.au
Name of Joint/Co-Supervisor: Jelena Rnjak-Kovacina
Email of Joint/Co-Supervisor: j.rnjak-kovacina@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Biomaterials and Tissue Engineering
Applicable to other Engineering
schools/disciplines:
Chemical Engineering
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Abstract: Donor supply of blood stem cells for blood production and transplantation in humans is limited. We have developed a method to generated adult-like blood stem cells from embryonic stem cell lines using microfluidics. These cells arise from arterial precursor cells that grow in culture when exposed to fluid shear stress. This project will investigate 'industrialisation' of this process so that blood can be derived from stem cell lines, rather than relying on blood donors.
Research Environment: Our lab has expertise in a) lab-on-a-chip fabrication b) embryonic stem cell culture and differentiation and c) live cell imaging. The culture system has been developed by PhD student Jingjing Li. This project is funded by Stem Cells Australia.
Novelty and Contribution: This is the first description of a culture system that generates blood stem cells using fluid shear. There are many novel aspects of this project that require development including bioreactors for generating large numbers of blood stem cells, fluid dynamic modeling of microfluidic devices, and tracking development of blood stem cells by live cell imaging.
Expected Outcomes: a) Development of a bioreactor for generating large numbers of blood stem cells
b) Mapping of the differentiation pathway from embryonic stem cells to blood stem cells
c) Understanding the biochemical mechanism that transduces the shear stress signal
Reference Material Links: 1. Chen, H. and R.E. Nordon, Application of microfluidics to study stem cell dynamics, in Emerging Trends in Cell and Gene Therapy, M. Danquah and R.I. Mahato, Editors. 2013, Springer: New York. p. 705.
2. Chen, H., J. Cornwell, H. Zhang, T. Lim, R. Resurreccion, T. Port, G. Rosengarten, and R.E. Nordon, Cardiac-like flow generator for long-term imaging of endothelial cell responses to circulatory pulsatile flow at microscale. Lab on a Chip, 2013. 13(15): p. 2999-3007.
3. Ng, E.S., L. Azzola, et al, Differentiation of human embryonic stem cells to HOXA+ hemogenic vasculature that resembles the aorta-gonad-mesonephros. Nat Biotechnol, 2016. 34(11): p. 1168-1179.

Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Improving the accuracy of luminescence imaging for solar cell research and manufacturing.
Name of Supervisor: Dr David Payne
Email of Supervisor: d.n.payne@unsw.edu.au
Name of Joint/Co-Supervisor: Dr Mattias Juhl
Email of Joint/Co-Supervisor: mattias.juhl@unsw.edu.au
School: School of Photovoltaic and Renewable Energy Engineering
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Characterisation
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Photoluminescence and electroluminescence imaging is routinely used in photovoltaic research and industrial manufacturing as a means to extract useful material information for scientific insight as well as quality control. However, such imaging systems typically use silicon CCD cameras which are susceptible to a photon smearing effect at silicon luminescence wavelengths, leading to blurred and somewhat inaccurate images.
A small team of SPREE researchers are investigating methods to minimise this effect and improve image accuracy through point spread function deconvolution. A taste of research student working on this project will focus on quantifying the accuracy of several deconvolution algorithms and determine the impact of various sample and imager parameters. They will acquire hands on experience of luminescence based solar cell characterization, using a range of both commercial and unique custom tools available at UNSW.
Research Environment: The student will work closely with several postdoctoral research fellows to acquire data from a broad range of silicon based samples. Outside of the lab they will work on analysis of the data using custom software developed within SPREE, which they will have the opportunity to further develop.
Prior experience with Matlab or Python coding would be beneficial but is not essential, as developing these skills will be part of the project.
Novelty and Contribution: The outcomes of this work have significant potential to improve characterisation techniques used on a global scale throughout PV research and industry.
Expected Outcomes: Improvements in imaging accuracy and quantification of error are key outcomes of this work. If the project progresses well then these findings will be submitted for journal publication. Improvements and contributions to software development of tools which are widely used by PV researchers is another probable outcome.
Reference Material Links: A paper on this work so far can be found here:
http://dx.doi.org/10.1016/j.cpc.2017.02.012.

The following background courses would be helpful but not essential to this work.
SOLA3020
SOLA3507
SOLA5055
SOLA5508
SOLA5509
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Automation of data acquisition and control of an anaerobic, marine photobioreactor producing CH4
Name of Supervisor: Ivan Perez Wurfl
Email of Supervisor: ivanpw@unsw.edu.au
Name of Joint/Co-Supervisor: Tracey Yeung
Email of Joint/Co-Supervisor: tracey.yeung@unsw.edu.au
School: School of Photovoltaic and Renewable Energy Engineering
Faculty Research Area (Theme): Energy Systems, Renewable and Non-Renewable
School Research Area: Combustion and Biofuels
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Abstract: Bioenergy with carbon capture and storage (BECCS) represents a large group of negative emissions technologies that could play significant roles in strategies to combat global warming. We are working on a carbon-negative bioenergy system that produces methane by relying on processes naturally performed by various marine microorganisms. The system is a photobioreactor (PBR) containing microalgae, digesting bacteria and anaerobic, methanogenic archaea. The microalgae store light energy as biomass through photosynthesis, which is digested and decomposed by a large community of bacteria and archaea to produce methane. Additionally, some types of microalgae then capture and store carbon dioxide as calcium carbonate. By utilising these microorganisms, the PBR can produce methane while capturing and storing carbon in the atmosphere.

This project requires basic knowledge of data acquisition and measurements (SOLA2051) as well as basic knowledge in electronic circuits (ELEC1111 and SOLA2060). Familiarity with LabVIEW is a plus.
Research Environment: Working closely with one PhD student and supervisor in an exciting laboratory environment.
Novelty and Contribution: This will be the first automated bioreactor to be built at SPREE. It can potentially be a prototype that could be reproduced many times to run multiple experiments in parallel.
Expected Outcomes: The purpose of this project is to automate the control and data acquisition required to run long experiments (2-3 months) using this bioreactor. A prototype of the data acquisition system has been designed but has not been tested. Setting up the system will require ensuring that all sensors and actuators are properly working, connecting all parts together to work as a system and finally calibrating all sensors and actuators. In the last month of this project, the system should be tested using live microorganisms.
A successful system may be reproduced to multiple bioreactors. It is foreseen that the project will also entail designing and implementing a way to connect multiple systems like this in parallel.
Reference Material Links: Sabine Fuss et. al, Betting on negative emissions, Nature Climate Change 4, 850–853 (2014)
http://dx.doi.org/10.1038/nclimate2392
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: 5G IoT communications on software defined radio platform
Name of Supervisor: Jinhong Yuan
Email of Supervisor: j.yuan@unsw.edu.au
Name of Joint/Co-Supervisor: Raja Pillai
Email of Joint/Co-Supervisor: raja.pillai@ni.com
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Data and Mobile Networks
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: Next generation wireless and mobile communications require transmitting and receiving multimedia information with high quality and throughput. Current research shows that by employing multiple-transmit and multiple-receive antennas it is possible to achieve very high data rate transmission in rich scattering wireless environment. Signal processing and coding combined with multiple antennas are called space-time techniques. The project is proposed to design transmitted signals with the multiple access IoT techniques. The candidate will work with a senior researcher and postgraduate research students. The work could include modulation, wireless channel, receiver design, etc. Programming is required to implement the designed schemes. Further enquiry, please contact Prof. Jinhong Yuan at 9385 4244 or Jinhong@ee.unsw.edu.au.
Research Environment: The project will be conducted within Wireless Communications Research Group at EE&T, comprising a large number of researchers (junior and senior) and PhD students, who are very supportive and approachable. The candidate will work with senior researchers, engineers from NI, and postgraduate research students to develop advanced technologies.
Novelty and Contribution: The novelty of the project is its IoT and SDR design, which is different from the current designs for wireless communications. It is expected that the project will result in much improved data rates for next generation networks.
Expected Outcomes: The candidates are expected to demonstrate the designed multiple access IoT schemes on the NI SDR platform.
Reference Material Links: Contact J.Yuan@unsw.edu.au or raja.pillai@ni.com
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Building and Securing a Smart Campus
Name of Supervisor: Vijay Sivaraman
Email of Supervisor: vijay@unsw.edu.au
Name of Joint/Co-Supervisor: Hassan Habibi Gharakheili
Email of Joint/Co-Supervisor: h.habibi@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Data and Mobile Networks
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: This project is aimed at making UNSW a smart-campus. Example use-cases include instrumenting learning spaces to profile their usage, parking spaces to understand how much they are used, and vending machines to know how much foot-traffic goes in front of them. Students involved in this project can choose from a range of activities, including building new sensor boards, testing connectivity technologies like WiFi and LoraWAN, collecting and hosting the data, analysing the data to get insights, analysing the system for cyber-security threats, and developing voice-apps to interact with the data.
Research Environment: This project will be carried out in a vibrant research group that includes PhD and honours-thesis students at UNSW.
Novelty and Contribution: This project is at the cutting-edge of research into the Internet-of-Things (IoT) technology. Outputs from this project will not only be directly visible via web interfaces and apps, but will also help showcase UNSW as a leading “smart campus” in the world. Outcomes of this project are expected to lead to publications in international conferences and journals.
Expected Outcomes: Expected outcomes include the development of new hardware and software that enable use-cases around monitoring of learning spaces and parking spaces, as well as data visualisation, analysis, and security.
Reference Material Links: There are many articles on the Internet-of-Things in the popular press. Please refer to some of our recent conference papers related to IoT instrumentation, data collection, and security at http://www2.ee.unsw.edu.au/~vijay/publications.html to get an idea of the IoT related we have done in our research group, and how it relates to international work in this area.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: LoRa test-bed development and applications for agricultrural Internet of Things (IoT)
Name of Supervisor: Prof Jinhong Yuan
Email of Supervisor: j.yuan@unsw.edu.au
Name of Joint/Co-Supervisor: Dr Lei Yang
Email of Joint/Co-Supervisor: lei.yang3@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Data and Mobile Networks
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: Next generation wireless and mobile communications require transmitting and receiving multimedia information with high quality and throughput. In addition, it also needs to support and connect over 50 billion things around the world, which is called Internet of Things (IoT). This is a new paradigm of low power wide area networks (LPWAN), which has been developed with many vertical Internet of Things applications. The project is proposed to design IoT techniques. The candidate will work with Prof Yuan, DR Yang and senior researchers and postgraduate research students. The work could include test-bed development, evaluate its coverage & data rate, investigate its application for agricultural field, etc. Programming is required to implement the designed schemes. Further enquiry, please contact Prof. Jinhong Yuan at 9385 4244 or J.Yuan@unsw.edu.au or Dr Lei Yang at Lei.Yang3@unsw.edu.au


Research Environment: The taste of summer scholars will work closely with the researchers and students at the wireless communications group at EE&T at UNSW. Prof Yuan and Dr Yang will guide the student to set-up the experiments and to develop advanced technologies.
Novelty and Contribution: The novelty of the project is its IoT. LoRa is a LPWAN, a new paradigm of low power wide area networks Internet of Things applications. The main work of the project is to set-up a LoRa test-bed, evaluate its coverage & data rate, investigate its application for agricultural field.
Expected Outcomes: LoRa test-bed with LoRa access point (gateway) and power-saving endpoints. A potential research publication.
Reference Material Links: Contact J.Yuan@unsw.edu.au or Lei.Yang3@unsw.edu.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Human Insertion into HR
Name of Supervisor: John Page
Email of Supervisor: j.page@unsw.edu.au
Name of Joint/Co-Supervisor: Ali Ahmed
Email of Joint/Co-Supervisor: a.f.ahmed@unsw.edu.au
School: School of Mechanical and Manufacturing Engineering
Faculty Research Area (Theme): Immersive Systems and Virtual Reality
School Research Area: Design and Analysis
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: Though we have been very successful in generating fixed assets in a Virtual Environment there is still a lot of work required to insert human avatars or agents into those environments. For example we have generated a realistic virtual simulation of the university campus but have had more difficulty in seeding it with students and staff. While it is relatively easy to simulate one individual in a virtual environment it becomes much more difficult task to simulate groups of people. This work will investigate how this might be achieved using the many approaches that have been used for crowd simulation in simulation programs such as AnyLogic. Currently we use MakeHuman to generate human avatars or agents, Blender to insert them into a 3D environment and a number of motion capture data sets to animate them. They are then transferred to Unity for VR conditioning to apply to Oculus. Most of the software is either freeware of at least, in the case of Unity, free to the university. Of course during this exercise it might become apparent that an alternative suite of software is more appropriate to the task in which case we will modify our approach.
Research Environment: The student will work in the Simulation and Virtual Engineering (SAVE) Laboratory which is located in J17/311. The laboratory has al the hardware and software required to carry out the project and the student one briefed will have access to all the equipment within the laboratory. The post graduate students, who will assist, in the supervision are located inJ17/311 and will thus be available to assist in any way needed.
Novelty and Contribution: A cheap easily available VR capability is relatively new. Though we have been involved in applying virtual reality to engineering projects for a decade it is only over the past four or so years cheap easily available headsets have become available. When the Oculus Rift was first released it was a development kit followed by development kit two. We were involved as an early used of DK1 since when the software capability and our understanding has advanced dramatically. Never the less the human aspects of VR are still not fully developed and this is where we are concentrating our efforts.
Expected Outcomes: Because the taste of research program offers us opportunities to explore the research envelope in areas we would not have time to address it can lead to significant outcomes. In the past three students were included as authors on technical publications having provided substantial input. From the student own point of view there a few opportunities to work in a new disruptive technology, which VR simulation is, which provides insights into where the technology is going. It also provides the opportunity to work closely with our research students in an area interesting to them.
Reference Material Links: The most useful reference material prior to starting the project is the many instructional videos on YouTube that teach the use of MakeHuman, Blender and Unity.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Finding chlorine – an atomic-scale investigation underneath the shiny skin of steel.
Name of Supervisor: Patrick Burr
Email of Supervisor: p.burr@unsw.edu.au
Name of Joint/Co-Supervisor: Hassan Tahini
Email of Joint/Co-Supervisor: h.tahini@unsw.edu.au
School: School of Chemical Engineering
Faculty Research Area (Theme): Advanced Materials
School Research Area: Energy
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Civil & Environmental Engineering
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Photovoltaic and Renewable Energy Engineering
Sciences – Maths, Physics, Chemistry
Abstract: We are all familiar with rust – if you leave a piece of steel in the rain, or worse in sea water, it will slowly turn into a crumbly mess of rust. But a much more severe and stealthier problem for high end alloys is that of stress-corrosion-cracking, which happens under the shiny surface of stainless steel, and eats the material from the inside, leading to catastrophic failure of the component. The main culprit of this phenomenon is trace amounts of chlorine dissolved on the surface of the material. Simply touching a piece of steel with our bare hands will leave enough Cl on the surface to induce stress-corrosion-cracking if the conditions are right. Despite decades of investigation, there are still unsolved mysteries: Cl was though to penetrate the protective layer of stainless steel through atomically sharp cracks, but a few recent studies show that Cl is found ahead of the crack tip, suggesting it doesn’t need cracks to get into the material. So how does it get there?
The aim of this project is to use atomic scale simulations to investigate the diffusion process that makes this possible.
Research Environment: The student will interact with myself and other members of the research team on a daily basis, but he/she will have flexible working hours. We are a small, vibrant and welcoming team of both senior and junior researchers. There will be plenty of occasions for social gatherings (lunch/pub/etc) throughout the duration of the project. We are part of the Integrated Materials Design Centre (IMDC) and the Nuclear Research Group.
Most of the work will consist of computer simulations. The student will be given access to national supercomputers (MASSIVE-M2 and NCI-Raijin), and all the required software for visualization, computation and analysis.
Novelty and Contribution: The cost of corrosion is estimated at $1.1 trillion USD a year in the US alone. That’s 6% of the country’s GDP, going down the drain every year! And what’s more fascinating, is that we still don’t completely understand the whole picture. If we understand the mechanism behind this stealthy corrosion process, we can put in place methods to prevent it from happening, thus saving trillions of dollars every year.
The project may also open the doors to novel diffusion processes of amphoteric elements within oxides, with wide-spread repercussions throughout materials science and engineering.
Expected Outcomes: The results of the project will lead to publication in high impact peer-reviewed journals.
Milestones:
• A complete set of simulations of Cr2O3 crystal with Cl defect.
• Identified diffusion pathways for Cl in Cr2O3.
• Investigate effect of stain on the diffusivity of Cl in Cr2O3.
Reference Material Links: https://www.nace.org/Publications/Cost-of-Corrosion-Study/

Ryan et al., Nature 41 (2002), 770-774, doi:10.1038/415770a, http://www.nature.com/nature/journal/v415/n6873/abs/415770a.html
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Measuring from hearts, eyes and brains: Next generation optrode
Name of Supervisor: Prof Nigel Lovell
Email of Supervisor: N.Lovell@unsw.edu.au
Name of Joint/Co-Supervisor: Prof Francois Ladouceur
Email of Joint/Co-Supervisor: F.Ladouceur@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Implantable Bionics
Applicable to other Engineering
schools/disciplines:
Chemical Engineering
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: The proposed innovative step is to apply a new class of liquid crystals, so called deformed helix ferroelectric
(DHF) liquid crystal, to the task of sensing extracellular biopotentials. In response to an applied electrical field, we have shown that DHF crystals can modulate an externally applied polarised light source with extraordinary
sensitivity and linear response down to the microvolt range. Using this technology we will initially design and test a single optrode device on the bench, before in vitro testing and characterisation using two-photon microscopy. The final design will be a higher density sensor array using a fibre optic source and multiple optical couplers with validation using an in vitro cardiac tissue preparation.

Students could focus on either the sensor technology, optics, microscopy for testing the sensor, biological tissue preps used with the optrode, or on computational modelling of the optrode response.
Research Environment: GSBmE and EET have extensive facilities to conduct the device fab and testing. The project is funded by the Australian Research Council.
Novelty and Contribution: Successful realisation of the technologies described herein has the potential to revolutionise the way multi-site
recordings are made from excitable tissue preparations both in-vitro and in-situ. It will also herald the next
generation of brain-machine interfaces, alleviating many of the wiring and packaging issues associated with
existing devices, and introduce new ways to implant devices in the body for sensing and diagnostic purposes.
Expected Outcomes: Better understanding of how the optrode could be designed or tested.
Reference Material Links: Contact supervisors for more info.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Pushing heart pumps to the limit: increase heart pump output during exercise
Name of Supervisor: Michael Stevens
Email of Supervisor: michael.stevens@unsw.edu.au
Name of Joint/Co-Supervisor: Nigel Lovell
Email of Joint/Co-Supervisor: n.lovell@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Implantable Bionics
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Excited about autonomous systems? Interested in an exciting medical application for your engineering knowledge? Or perhaps you want to improve the quality of life of the tens of thousands of people who suffer heart disease? If so, read on...

Heart disease is one of the biggest killers in the developed world. The ideal treatment is a heart transplant, however the demand for donor hearts greatly outweighs supply.

Implantable heart pumps give these patients a chance at a life whilst they await a heart transplant. Their success has led to over 15 000 implantations worldwide, in both the young and the elderly. The next stage of research is to improve the quality of life of patients supported by heart pumps. One shortcoming of these devices is that they lack an automatic response to changes in cardiac demand.

In this project, you will investigate methods of improving the interaction between the human body and the implantable heart pump, through one of the following ways

- Developing a control system to automatically adjust speed based on changes in cardiac demand
- Develop methods of estimating blood pressure and flow using the current and power signals from the pump, which can then be used as feedback to a control system
- Using a computer simulation, investigate the effect of the pump connectors on pump flow in exercise.
Research Environment: The student will work at the Graduate School of Biomedical Engineering at UNSW. Our small but motivated team consists of a number of PhD students, supervised by one post-doc and a senior academic. You may also spend time at the Victor Chang institute performing experiments.
Novelty and Contribution: The novelty of this project is either a new method of controlling heart pumps, an improved estimation algorithm or quantification of the relationship between pump connectors and exercise flow rates. This contribution will eventually lead to more positive outcomes for patients.
Expected Outcomes: The outcome will be a fully tuned control system that can be tested using a bench top apparatus.
Reference Material Links: http://www.ncbi.nlm.nih.gov/pubmed/21665512
https://www.ncbi.nlm.nih.gov/pubmed/?term=Physiological+control+of+dual+rotary+pumps+as+a+biventricular+assist+device+using+a+master%2Fslave+approach
https://www.youtube.com/watch?v=EmFNvjmAmh0&t=7s
https://www.youtube.com/watch?v=KsAf-tMmpyg
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: The student may perform experiments at the Garvan Institute, near St Vincent's Hospital.

[Back to Top]


Project Title: Machine learning in Speech Processing
Name of Supervisor: Dr. Vidhyasaharan Sethu
Email of Supervisor: v.sethu@unsw.edu.au
Name of Joint/Co-Supervisor: A/Prof. Julien Epps
Email of Joint/Co-Supervisor: j.epps@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Signal Processing & Control
School Research Area: Multimedia Signal Processing
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Apart from the linguistic content (what is being said), a variety of other information, referred to as paralinguistic information, can be inferred from speech. Examples include the identity of the speaker, their gender, and their emotional and mental states. The development of automatic systems that can make these inferences are highly active areas of research and most of these are relatively recent endeavours. A major hurdle faced by these systems is that the collection and labelling of training data are both extremely hard. Consequently the datasets used are extremely small, which can lead to unreliable models.

The aim of this project is the preliminary investigation of the viability of systems that can make inferences within a Bayesian statistical framework with the specific aim of accounting for uncertainty due to small datasets in order to avoid sub-optimal inferences. If successful, this could be the first step in the path leading to computing systems that are able to recognise humans’ mental and emotional states in a way that only humans can at present.

At the completion of the project, the student will have a good understanding of machine learning/pattern classification and statistical modelling. In addition they will have significantly strengthened their technical/research skill as well as programming skills in MATLAB (and optionally C/C++). This project can be undertaken by individual students have an interest in applied mathematics and/or machine learning and have strong programming skills.
Research Environment: The work environment will be within the speech and audio processing laboratories of the School of Electrical Engineering and Telecommunications. In this laboratory there are 8 PhD students working on speaker verification, language identification, speaker recognition, emotion detection and speech enhancement.
Novelty and Contribution: Machine learning in speech based systems is a massive field of research and development with many large firms (such as Google, Apple, Samsung and Microsoft) developing their own in house systems for a number of speech based applications. Skills developed over the course of this project can be expected to be very valuable in this field. Also easy to use code/toolboxes for Bayesian inference that may be developed as part of this project can be expected to attract a good deal of interest in the speech research community.
Expected Outcomes: MATLAB (and optionally C/C++ and CUDA) code, a functional MATLAB toolkit for Bayesian machine learning, a written report, and code documentation. If the topic were extended into an honours thesis, more would be possible.
Reference Material Links: Contact v.sethu@unsw.edu.au or drop by MSEB649 to discuss the topic. (Enquiries are encouraged). A video lecture series that talk about Bayesian methods and pattern recognition can be found at:
http://videolectures.net/course_information_theory_pattern_recognition/

Suggested reading: http://www.inference.phy.cam.ac.uk/mackay/itila/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Quantum Communications with the Chinese Quantum Satellite
Name of Supervisor: Robert Malaney
Email of Supervisor: r.malaney@unsw.edu.au
Name of Joint/Co-Supervisor: Xiaoyu Ai
Email of Joint/Co-Supervisor: x.ai@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Network Privacy, Security and Quantum Telecommunications
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Quantum Communications and Quantum Networks are anticipated to be the core networking technologies of the 21st century. In fact these communication systems have already appeared in the commercial world in many variations. However, just today (16/06/17) the results from the World's first quantum-enabled satellite (launched last year by China) have appeared in Science. This represents a gigantic step forward for quantum communications and takes the so-called quantum internet (a collection of quantum computers talking to each other via space) a step closer to reality. In this project you will investigate realistic communication scenarios that can arise from the use of this new satellite, including unconditionally secure communications, quantum teleportation, and unconditional location verification.
Research Environment: The School of Electrical Engineering and Telecommunications, University of New South Wales (UNSW), is one of Australia's leading research and teaching schools in the area of Telecommunications. UNSW has a world class intellectual environment. The School has excellent facilities with extensive computational and research infrastructure and provides strong research support in wireless and quantum communications. More specifically, the school has well-known experts in the areas of channel coding, MIMO signal processing, physical-layer security, location technologies, quantum communications and quantum computing, regularly publishing in high impact journals and prestigious international conferences in these areas. In this project you will work closely with our research staff and PhD students working in the area of quantum communications.
Novelty and Contribution: The project will involve detailed numerical simulations of realistic communication scenarios that model as closely as possible the communication set up being used by the Chinese quantum satellite. Once the current experimental results have been reproduced you will be tasked with exploring new scenarios and set-ups that could be used for future use. These include teleportation, location verification and the use of hybrid quantum states as a means of enhancing communications.
Expected Outcomes: A full blown simulation model in Matlab that encompasses use of control theory as a means of switching between discrete quantum states and continuous quantum states - based on feedback from laser measurements of the atmospheric turbulence between low-Earth orbit and ground. A written report will also be required at the end of the project.
Reference Material Links: ABC News:
http://www.abc.net.au/news/science/2017-06-16/chinese-satellite-breaks-quantum-entanglement-distance-record/8620240

Science: Satellite-based entanglement distribution over 1200 kilometers:
http://science.sciencemag.org/content/356/6343/1140
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Enhancing Building Information Modelling with Supply Chain Data
Name of Supervisor: Dr. Mojtaba Maghrebi
Email of Supervisor: maghrebi@unsw.edu.au
Name of Joint/Co-Supervisor: Prof. Travis Waller
Email of Joint/Co-Supervisor: s.waller@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Management
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Mining Engineering
Abstract: This study is about enhancing Building Information Modelling (BIM) with supply chain data in order to minimise the delays caused by lack of materials on construction sites especially for those projects that have international suppliers and some of their critical materials are supplied from overseas. In this study the supply chain process are automatically tracked during the projects not only just before commencement of tasks.
Research Environment: office
Novelty and Contribution: Developing a tool that can minimise the delays caused by lack of materials on sites.
Expected Outcomes: A decision support system or a plug-in/app
Reference Material Links: https://www.degruyter.com/downloadpdf/j/jbe.2013.1.issue-2/jbe-2013-0009/jbe-2013-0009.pdf
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Geospatial Mapping and Navigation Requirements for Autonomous Driving
Name of Supervisor: Jinling Wang
Email of Supervisor: jinling.wang@unsw.edu.au
Name of Joint/Co-Supervisor: Shuraun Zheng
Email of Joint/Co-Supervisor: zhshran@gmail.com
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Spatial Information Systems and Positioning
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Mining Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Google has been investigating and testing autonomous cars or driverless cars for many years which has attracted huge public attention. Recently, major car manufacturers have also announced that they will produce autonomous driving technology as an option for next generation cars by 2020. Such technological advancements will make a significant impact on future transport engineering in general. This project will investigate the critical issues in geospatial mapping and navigation for Autonomous Driving.
Research Environment: Geospatial mapping and navigation research activities at UNSW are carried out with modern Global Navigation Satellite Systems (GNSS), inertial navigation, Light Detection and Ranging (LiDAR) and vision sensors. Since 2004, a total of 7 PhD students under supervision of A/Prof Jinling Wang have won prestigious Student Paper Awards at the US Institute of Navigation (ION) GNSS+ meeting, which is the world’s premier international technical meeting and showcase for satellite navigation and positioning technology.
Novelty and Contribution: As a critical part of next generation transport system, navigation technology will paly an important role in autonomous driving in an integrated transport system. For example, autonomous driving will require a dedicated 3D map. Indeed, optimized map data are needed for orientation and positioning in autonomous driving. At the same time, navigation sub-system should be able to meet certain integrity requirements to address transport safety and liability concerns related to autonomous driving. These issues are not yet solved.
Expected Outcomes: This project aims to investigate major aspects of the mapping and navigational requirements for autonomous driving in future integrated transport systems. The expected outcomes include:
a)Structure of the High Definition Maps for highly automated driving;
b)Reliability measures of navigation for highly automated driving;
c)Statistical analysis of High Definition Maps and Navigation
Reference Material Links: http://www.austroads.com.au/drivers-vehicles/connected-and-automated-vehicles/projects;

Emerging Digital Mapping Requirements for C-ITS:
https://www.onlinepublications.austroads.com.au/items/AP-R432-13

Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Integrating Supply Information with Construction Planning
Name of Supervisor: Dr. Mojtaba Maghrebi
Email of Supervisor: maghrebi@unsw.edu.au
Name of Joint/Co-Supervisor: Prof. Travis Waller
Email of Joint/Co-Supervisor: s.waller@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Resources Engineering
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Abstract: One of the main reasons for delays in construction projects has its source on lack of materials on site on time. Unexpected changes on the schedule during the construction progress may cause changes in the dates materials should be ready to be used on site. The consequences for that can be massive, including the idle time of crew, the waste of money and resources and contractual issues with the project sponsors. This paper proposes an automated solution to minimise this this kind of delays in construction sites by integrating the supply chain process with Building Information Modelling(BIM). In details, all the supply chain characteristics are embedded in the schedule planner module and any change in the project schedule would also be applied into the supply chain planner module. Using feed forward controlling method, BIM would be used to simulate project sequencing and the tasks are arranged/rearranged until the schedule is considered satisfactory by the project manager. Only after the schedule plan is improved, the tasks may begin. This process is done automatically by BIM and it includes the supply chain requirements. So, not just a toilsome work is avoided, but also a more precise d grounded schedule planner is provided.
Research Environment: School of Civil and Environmental Engineering
Novelty and Contribution: Introducing a platform that can consider both supply and construction process in order to avoid delays caused by lack of materials.
Expected Outcomes: Reducing the construction delays
Reference Material Links: Maghrebi M; Sammut C; Waller S, 2013, 'Integrated Building Information Modeling (BIM) with Supply Chain and Feed-Forward Control', YBL Journal of Built Environment, vol. 1, no. 2, pp. 25 - 34, http://dx.doi.org/10.2478/jbe-2013-0009
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Logistics in humanitarian emergency operations
Name of Supervisor: Hanna Grzybowska
Email of Supervisor: h.grzybowska@unsw.edu.au
Name of Joint/Co-Supervisor: Lauren Gardner
Email of Joint/Co-Supervisor: l.gardner@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Management
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: The current project addresses the topic of logistics in humanitarian emergency operations. Conflicts and natural disasters around the world have been causing an influx of thousands of refugees into neighbouring countries which oftentimes get overwhelmed while their logistics capabilities get challenged. The importance of logistics in the provision of aid to disaster survivors as well as its role in carrying out other disaster management operational activities is undeniable. The design of logistics systems and the importance of coordination in carrying out logistics operations efficiently and effectively are critical. The aim of this project is to examine the processes involved in emergency relief and highlight the relationships between the participating bodies. The final goal is to define a comprehensive logistics model for humanitarian emergency operations.
Research Environment: Dr Grzybowska and Dr Gardner are located at the Research Centre for Integrated Transport Innovation (rCITI) at the School of Civil and Environmental Engineering
Novelty and Contribution: This study contributes to the state-of-the-art by providing an all-encompassing formulation of the humanitarian logistics model and identifying aspects that could be improved
Expected Outcomes: • Documented literature review,
• A model of the humanitarian logistics operations.
Reference Material Links: - Pettit, S.J. and Beresford, A.K., 2005. “Emergency relief logistics: an evaluation of military, non-military and composite response models”. International Journal of Logistics: Research and Applications, 8(4), pp.313-331.
- Chandes, J. and Paché, G., 2010. “Investigating humanitarian logistics issues: from operations management to strategic action.” Journal of Manufacturing Technology Management, 21(3), pp.320-340.
- Kovács, G. and Spens, K., 2009. “Identifying challenges in humanitarian logistics.” International Journal of Physical Distribution & Logistics Management, 39(6), pp.506-528.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Modelling vehicle behaviour at an un-signalised intersection
Name of Supervisor: Hanna Grzybowska
Email of Supervisor: h.grzybowska@unsw.edu.au
Name of Joint/Co-Supervisor: Travis S. Waller
Email of Joint/Co-Supervisor: s.waller@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Programming Languages and Software Engineering
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: This research project addresses the need for modelling vehicle behaviour at an un-signalised intersection. In this case vehicle behaviour is opposed to the one operating on a priority basis; i.e., when some vehicles are given right and/or indicated when to move by traffic lights, while the other vehicles adjust. Recent attempts have been made to emulate the interactions between vehicles at an un-signalised intersection through microsimulation approaches. However, the scope of these models has been confined to isolated intersections and short highways. For example, Makarem, et al. (2011) were able to simulate an un-signalised intersection in Aimsun and Matlab. The key parameters they used to assess the intersection were vehicle average speed, number of stops, vehicle throughput, fuel consumption and vehicle emissions. The challenge of the current project is to demonstrate the validity of the existing approaches on a larger scale and potentially extend them if required. The project is computationally demanding and requires advanced knowledge of: Python, Aimsun, and Matlab.
Research Environment: Dr Grzybowska and Prof Waller are located at the Research Centre for Integrated Transport Innovation (rCITI) at the School of Civil and Environmental Engineering
Novelty and Contribution: This study contributes to the state-of-the-art by identifying extensions to the existing models resulting from consideration of larger scale of the problem.
Expected Outcomes: • Documented literature review including a mathematical model formulation modelling the behaviour of vehicles at an un-signalised intersection,
• Implementation of the model in microsimulation environment using AIMSUN (requires development of a plugin in Python and data analysis using Matlab)
• Testing of the model in the microsimulation environment,
• Final summary report including results from testing.
Reference Material Links: Makarem, L., Pham, M.H., Dumont, A.G. and Gillet, D., 2012. “Microsimulation Modeling of Coordination of Automated
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: The impact of satellite data quality on hydro-ecological models
Name of Supervisor: Dr Lucy Marshall
Email of Supervisor: lucy.marshall@unsw.edu.au
Name of Joint/Co-Supervisor: Yating Tang
Email of Joint/Co-Supervisor: yating.tang@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Water and Wastewater Engineering
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Hydrologic models are powerful tools for water resources management and analysis. These models are particularly important for understanding how catchments will change under future climates, and for examining feedbacks between water and vegetation dynamics. The advent of satellite observational systems has vastly increased the usefulness of these models across the globe. However, satellite data can have significant errors or uncertainties.

This project is aimed at developing a numerical approach to ecohydrologic model optimization, taking into account the quality of satellite data. The project will examine multi-objective model calibration across multiple Australian catchments, and determine the extent to which data quality affects model outcomes.
Research Environment: The research will take place in Water Research Centre (WRC), an internationally renowned university centre that provides multidisciplinary research in water resources, engineering, management and the development of tools for environmental management and sustainability. The successful applicant will join a research team within the WRC comprising junior research staff and postgraduate students, and will additionally be mentored by a senior researcher.
Novelty and Contribution: To date, there is no established method for taking into account satellite data quality in ecohydrologic model simulations. This project will provide computational methods for optimizing ecohydrologic models, and will help develop better understanding of the impact of satellite data for models across multiple regions in Australia.
Expected Outcomes: By the end of the project, it is expected that the successful applicant will
(1) acquire in-depth knowledge of hydrologic models, ecohydrologic processes, and satellite data retrieval and processing;
(2) develop good programming and optimization skills;
(3) contribute to a research report or journal article describing the study methods and results.
Reference Material Links: http://www.hydrology.unsw.edu.au
http://www.sciencedirect.com/science/article/pii/S0022169415006757
http://www.hydrol-earth-syst-sci-discuss.net/hess-2016-573/
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: VR Driller: Virtual-Reality Deep Drilling Simulator with head mounted displays and a hand controller
Name of Supervisor: Stuart Walsh
Email of Supervisor: stuart.walsh@unsw.edu.au
Name of Joint/Co-Supervisor: James Tibbett
Email of Joint/Co-Supervisor: james.tibbett@unsw.edu.au
School: School of Petroleum Engineering
Faculty Research Area (Theme): Immersive Systems and Virtual Reality
Applicable to other Engineering
schools/disciplines:
Civil & Environmental Engineering
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mining Engineering
Abstract: Today's oil and gas wells are complex multi-million dollar affairs, requiring advanced equipment and specialist knowledge to drill and complete. In particular, use of horizontal drilling and stimulation techniques in unconventional resources means modern Petroleum Engineers need an intuitive understanding of the 3D arrangement of wells, reservoirs, and fractures when planning and implementing these operations.

Virtually reality provides an ideal platform for demonstrating modern drilling and completion techniques to up-and-coming engineers. Through a virtual environment, students can experience being on the drill platform, direct its operation, and see the effect of their decisions at depth.

The potential users of such a tool extend beyond engineering students. In recent years, public concerns regarding hydraulic fracturing have grown; in large part due to a lack of understanding concerning the nature of these operations. A VR platform provides an ideal environment to demonstrate the realities of modern drilling operations.

This project will develop an immersive educational tool to demonstrate drilling at depth using modern techniques. The virtual reality tool will help to educate UNSW students from Petroleum and Geothermal Engineering on drilling and completion, and provide a platform to inform the public on matters concerning modern drilling operations.
Research Environment: The Taste of Research candidate will collaborate with a team from the School of Petroleum Engineering and the School of Mining Engineering. The candidate will work closely with staff from both schools to plan the immersive VR experience. Through this Taste of Research project, the candidate will gain hands on training in programming and operating the School of Mining Engineering's Virtual reality simulator.
Novelty and Contribution: This project will help to develop a novel informational and educational experience for both students at UNSW and the public at large. It will provide a platform that can be used to help educate the next generation of petroleum and geothermal engineering students on the use of modern drilling and completion techniques. In addition it will provide a tool that can be used to inform the public on matters concerning modern petroleum engineering operations.
Expected Outcomes: The Taste of Research project will result in a virtual reality simulation that can be used to demonstrate modern drilling and completion techniques. The Virtual Reality experience will be provided through the use of Head Mounted Displays in the School of Mining Engineering's immersive VR Suite. Objects within the virtual reality environment will be manipulated with a Hand Controller.
Reference Material Links: Information on the Virtual Reality Simulator can be found at:
https://www.engineering.unsw.edu.au/mining-engineering/research/research-groups/innovative-learning-and-teaching/virtual-reality-simulator-a-world-first
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Workforce Scheduling and Routing Problem
Name of Supervisor: Hanna Grzybowska
Email of Supervisor: h.grzybowska@unsw.edu.au
Name of Joint/Co-Supervisor: Travis S. Waller
Email of Joint/Co-Supervisor: s.waller@unsw.edu.au
School: School of Civil and Environmental Engineering
Faculty Research Area (Theme): Management
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: This research project addresses Workforce Scheduling and Routing Problem (WSRP) encountered in human resources operations. Its goal is to optimally route and schedule a limited set of staff who serve a number of requests for service, while considering a set of pre-defined side constraints. The side constraints might include: definitions of time windows, customer and staff preferences and/or fairness, skills, equipment, collaborations, job emergency levels, compulsory rest breaks, and additional restrictions provided by the unions and/or specific to the profession. The problem is relevant and challenging to many companies providing maintenance operation services on locations such as in telecommunications, industrial machinery, public utilities, and health care industry and charity operations. It is also an interesting research problem. The challenge is to define a rich formulation of the WSRP including all the possible constraints and variations of the problem.
Research Environment: Dr Grzybowska and Prof Waller are located at the Research Centre for Integrated Transport Innovation (rCITI) at the School of Civil and Environmental Engineering
Novelty and Contribution: This study contributes to the state-of-the-art by providing an all-encompassing rich formulation of the Workforce Scheduling and Routing Problem and a design for a comprehensive problem solving method.
Expected Outcomes: • Documented literature review,
•A mathematical model for the rich Workforce Scheduling and Routing Problem.
•A design and development of a heuristic approach for problem solving.
Reference Material Links: Grzybowska, H., Gretton, C., Kilby, P. and Waller, S.T., 2015. “Decision Support System for a Real-Time Field Service Engineer Scheduling Problem with Emergencies and Collaborations”. Transportation Research Record: Journal of the Transportation Research Board, (2497), pp.117-123.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Tungsten boride: a new material for fusion.
Name of Supervisor: Patrick Burr
Email of Supervisor: p.burr@unsw.edu.au
Name of Joint/Co-Supervisor: Judy Hart
Email of Joint/Co-Supervisor: j.hart@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Advanced Materials
School Research Area: Nuclear Engineering
Applicable to other Engineering
schools/disciplines:
Biomedical Engineering
Chemical Engineering
Computer Science & Engineering
Mechanical & Manufacturing Engineering
Photovoltaic and Renewable Energy Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Fusion power is a potential avenue of clean, carbon-free, renewable energy source. Fusion reactors they have been around for decades, but they are still not commercially viable. The main reason for this is that we are yet to develop materials that can withstand the extreme condition of fusion plasmas over the lifetime of the reactor (~60-100 years). The plasma reaches temperatures above 100 million ºC, contains fast moving particles (at speeds of hundreds of km/s), and produces a steady flow of high energy neutrons and gamma rays. Meanwhile, at the very centre of the fusion chamber lie superconducting magnets, which are kept at –269 ºC, and are very sensitive to radiation damage from neutrons and gamma.
The challenge is to develop shielding materials with ultra-high melting point, extreme hardness, capable of stopping neutron in a short distance and with exceptional tolerance to radiation damage.
Tungsten carbide is one of the main candidate materials and our research has shown that it has a good combination of properties, but a trade-off is made between wall thickness and neutron/gamma stopping stopping-power. The aim of this project is to improve on the previous work by substituting carbon for boron, which has a much stronger tendency to absorb neutrons. The student will perform atomic-scale simulations of a range of tungsten borides (WB2, WB, W2B, W2B5, and WB4) and will also consider carbide-boride solid solutions and composite materials.
Research Environment: The student will interact with myself and other members of the research team on a daily basis, but he/she will have flexible working hours. We are a small, vibrant and welcoming team of both senior and junior researchers. There will be plenty of occasions for social gatherings (lunch/pub/etc) throughout the duration of the project. We are part of the Integrated Materials Design Centre (IMDC) and the Nuclear Research Group.
Most of the work will consist of computer simulations. The student will be given access to national supercomputers (MASSIVE-M2 and NCI-Raijin), and all the required software for visualization, computation and analysis.
Novelty and Contribution: This project will inform the fusion community on the viability of tungsten carbide/boride materials for shielding of critical components. The outcome of this project will directly influence experimental efforts on martials design. Ultimately, if successful, the project will enable improved performance in fusion reactors.
Expected Outcomes: The results of the project will lead to publication in high impact peer-reviewed journals.
Milestones:
• Simulations of defect formation energies and defect clustering in W borides
• Calculations of diffusion coefficients for defects in W borides
• Solubility and partitioning of transmutation elements of W
• Identification of viable carbide-boride solid solutions
Reference Material Links: P. Träskelin, C. Björkas, N. Juslin, K. Vörtler, and K. Nordlund, Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 257, 614 (2007).
A. Lasa, C. Björkas, K. Vörtler, and K. Nordlund, J. Nucl. Mater. 429, 284 (2012).
K. Vörtler and K. Nordlund, J. Phys. Chem. C 114, 5382 (2010).
X. S. Kong, Y. W. You, J. H. Xia, C. S. Liu, Q. F. Fang, G. N. Luo, and Q. Y. Huang, J. Nucl. Mater. 406, 323 (2010).
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: Depending on the results of the project, there may be scope for some fabrication and experimentation to test the finding of the simulations. These may be carried out using facilities at UNSW and ANSTO (Australian Nuclear Science and Technology Organisation).

[Back to Top]


Project Title: A thermal imaging sensor for unobtrusive monitoring of falls in elderly people
Name of Supervisor: Nigel Lovell
Email of Supervisor: n.lovell@unsw.edu.au
Name of Joint/Co-Supervisor: Michael Stevens
Email of Joint/Co-Supervisor: michael.stevens@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Physiological Measurement
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Falls are a major health problem facing older people, with an estimation that one out of every three people over the age of 65 fall once per year. Given that over half of those who fall cannot get up without assistance, it is likely that a fall can develop into a “long-lie” scenario. It is estimated that approximately half of those who experience a long lie will die within 6 months. It follows that rapid detection of falls can reduce the likelihood of a long-lie through faster response from caregivers.

In this project, you will have the exciting opportunity to contribute to the design of a very different type of fall detector to that commonly used. Instead of a wearable device we will work with a start-up company (Calumino) that are designing a thermal imaging sensor. This sensor will be connected to the cloud as an Internet-of-Things device and images captured in simulated homes of patients will be analysed to detect the presence of the person within a home environment, and by way of image processing, whether the person has fallen and is immobile.
Research Environment: The student will work at the Graduate School of Biomedical Engineering at UNSW. Our team consists of a number of PhD students and post-doctoral researchers from various backgrounds (electrical, mechanical and biomedical engineering), led by 2 senior academics.
Novelty and Contribution: The aim of this project is to develop a falls monitoring technology that is unobtrusive and does not require the person to remember to wear a device. No such projects using thermal imaging have been done in the past.
Expected Outcomes: The student will produce a prototype of the design incorporating their modifications, and assess the feasibility and effectiveness of their modifications.
Reference Material Links: Wang C, Narayanan MR, Lord SR, Redmond SJ, Lovell NH (2014) A low-power fall detection algorithm based on triaxial acceleration and barometric pressure. Conf Proc IEEE Eng Med Biol Soc. 570–573.

See supervisors for more references
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: May involve visits to the Technology Park to discuss designs with the start-up company Calumino.

[Back to Top]


Project Title: Development of a device for rapid detection of falls in elderly people
Name of Supervisor: Michael Stevens
Email of Supervisor: michael.stevens@unsw.edu.au
Name of Joint/Co-Supervisor: Nigel Lovell
Email of Joint/Co-Supervisor: n.lovell@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Health & Medical Technologies
School Research Area: Physiological Measurement
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Are you looking for a challenging multi-disciplinary engineering problem with a real-world application? One that can have direct impact on the health of elderly Australians?

Falls are a major health problem facing older people, with an estimation that one out of every three people over the age of 65 fall once per year. Given that over half of those who fall cannot get up without assistance, it is likely that a fall can develop into a “long-lie” scenario. It is estimated that approximately half of those who experience a long lie will die within 6 months. It follows that rapid detection of falls can reduce the likelihood of a long-lie through faster response from caregivers.

In this project, you will have the exciting opportunity to contribute to the design of a revolutionary new falls detector. Given the multi-disciplinary nature of the problem, the project will be tailored to your engineering major. As such you may find yourself modifying the current device in one (or more!) of the following ways:
• Designing a new enclosure to be robust, easy to use, manufacturable and waterproof.
• Reducing the power consumption of the falls detection software.
• Utilising renewable sources of energy to power wearable devices.
• Identifying new features of falls that can be used to improve algorithm accuracy.
Research Environment: The student will work at the Graduate School of Biomedical Engineering at UNSW. Our team consists of a number of PhD students and post-doctoral researchers from various backgrounds (electrical, mechanical and biomedical engineering), led by 2 senior academics.
Novelty and Contribution: The aim of this project is to develop the most accurate and least power-hungry falls detector on the market. This device can then be deployed in the real world, first in aged-care facilities, followed by the homes of individuals.
Expected Outcomes: The student will produce a prototype of the design incorporating their modifications, and assess the feasibility and effectiveness of their modifications.
Reference Material Links: Bianchi F, Redmond SJ, Narayanan MR, Cerutti S, Lovell NH (2010) Barometric pressure and triaxial accelerometry-based falls event detection. IEEE Trans. Neural Syst. Rehabil. Eng. 18: 619–627.
Wang C, Narayanan MR, Lord SR, Redmond SJ, Lovell NH (2014) A low-power fall detection algorithm based on triaxial acceleration and barometric pressure. Conf Proc IEEE Eng Med Biol Soc. 570–573.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Model and design of a rapid thermal processing reactor
Name of Supervisor: Yansong Shen
Email of Supervisor: ys.shen@unsw.edu.au
Name of Joint/Co-Supervisor: Yuting Zhou
Email of Joint/Co-Supervisor: y.zhuo@student.unsw.edu.au
School: School of Chemical Engineering
Faculty Research Area (Theme): Fluid Dynamics and Thermal Engineering
School Research Area: Process and Products
Applicable to other Engineering
schools/disciplines:
Civil & Environmental Engineering
Computer Science & Engineering
Mechanical & Manufacturing Engineering
Photovoltaic and Renewable Energy Engineering
Abstract: Rapid Thermal Processing (RTP) refers to a semiconductor manufacturing process which heats silicon wafers to high temperatures (over 1,000 °C) on a timescale of several seconds or less. During cooling, however, wafer temperatures must be brought down slowly to prevent dislocations and wafer breakage due to thermal shock. Such rapid heating rates are often attained by high-intensity lamps or lasers. These processes are used for a wide variety of applications in semiconductor manufacturing including photovoltaic solar cells. This project aims to model the multiphase flow and phase changes in an RTP reactor for photovoltaic solar cell manufacturing.
Research Environment: The project will be conducted in ProMO research group (Process Modelling and Optimisation) in School of Chemical Engineering. The research at ProMO is powered by its hardware facility including advanced GPU-equipped HPC clusters and software resources including state-of-the-art numerical techniques collected from previous research projects e.g. continuum-discrete coupling approach for fluid-solid reactive systems.
Novelty and Contribution: The project represents a novel scientific contribution. A new CFD-based computer model will be developed to describe the multiphase flow in the RTP reactor. The model will then be used for design the new RTP reactor for manufacturing solar cell and other semiconductor-based products.
Expected Outcomes: A paper will be published in a learnt journal in the field.
An advanced CFD-based computer model will be developed.
Reference Material Links: https://research.unsw.edu.au/people/dr-yansong-shen
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Simulation Studies on Distributed Control of Renewable Energy Storage
Name of Supervisor: Prof. Jie Bao
Email of Supervisor: j.bao@unsw.edu.au
Name of Joint/Co-Supervisor: Dr Ruigang Wang
Email of Joint/Co-Supervisor: ruigang.wang@unsw.edu.au
School: School of Chemical Engineering
Faculty Research Area (Theme): Energy Systems, Renewable and Non-Renewable
School Research Area: Process and Products
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Photovoltaic and Renewable Energy Engineering
Abstract: The ever-increasing integration of distributed renewable energy generation sources with the electricity grid reduces our reliance on fossil fuels and carbon emissions but also presents risks to the grid’s stable and reliable operation due to intermittent nature of such sources.

This project will be a part of research effort to develop some key technologies of battery energy storage and control to address the above issues and help defer the investment for the augmentation of the transmission and distribution networks. The idea is to implement a network of distributed battery energy storage systems close to the users. The local battery system is used to store the excess power from the renewable energy sources, perform peak shaving (i.e., to charge the battery when the electricity price is cheap/power demand is low) and redistribute power to minimize the difference between total power supply and demand within a microgrid. The battery systems are also used to improve voltage stability and power quality by coordinated power sharing (e.g., to fast inject power to the microgrid when grid voltage is low). This project will develop an approach for controlling and coordinating the battery systems, which is the key problem in this application.
Research Environment: You will work with a research team with one academic supervisor, one research fellow and two PhD students.
Novelty and Contribution: You will join the Process Control Research Group, School of Chemical Engineering to develop a novel distributed control approach for controlling and coordinating distributed energy storage systems. This approach coordinates individual battery controllers to achieve best economic benefit for individual users while maintain microgrid-wide performance to limit oscillations of total energy demand and supply from a microgrid to the main grid.
Expected Outcomes: You will implement the control algorithms for the distributed (in individual autonomous controllers) developed by the Process Control Group to perform simulation studies.

The expected outcomes include the analysis of the effectiveness of the proposed networked control approach and its possible weakness.
Reference Material Links: Zhang X.N., Bao J., Wang R.G., Zheng C.X., Skyllas-Kazacos M. (2017) Dissipativity Based Distributed Economic Model Predictive Control for Residential Microgrid with Renewable Energy Generation and Battery Energy Storage. Renewable Energy 100: 18–34.

Tippett M.J. and Bao J. (2014) Dissipative Control of Plant-wide Process Systems using Dynamic Supply Rates. Automatica 50 (1): 44-52.

Tippett M. and Bao J. (2013) Distributed Model Predictive Control Based on Dissipativity. AIChE Journal 59: 787–804.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Improving quantum bit readout and control fidelity using MATLAB
Name of Supervisor: Prof. Andrew Dzurak
Email of Supervisor: a.dzurak@unsw.edu.au
Name of Joint/Co-Supervisor: Dr. Henry Yang
Email of Joint/Co-Supervisor: henry.yang@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): MEMS, Micro & Nano Technologies
School Research Area: Quantum Computing and Microelectronics
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Our research group at UNSW has developed a new type of quantum computing (QC) based on silicon CMOS technology that has gained great attention world-wide due to its compatibility with industrial CMOS manufacturing. In contrast to a classical computer, a quantum computer is composed of quantum bits (qubits), which can possess the on-off states of a regular bit, but also quantum superpositions of these “0” and “1” states. To operate a qubit, we communicate with the measurement instruments using MATLAB. The instruments provide the capability to readout and control our qubit states. Here, we aim to produce a software program which can assist in fine-tuning a multi-qubit device into its operation mode, and to readout the qubit states with high fidelity (or accuracy). Apart from that, it is also essential to achieve a high control fidelity on our qubit operation. This can be made possible by incorporating shaped pulses, simulated using MATLAB in our control pulses to manipulate the qubit state. These are two vital components for the operation of a future quantum computer, meeting error correction thresholds.
This Taste of Research student should possess good MATLAB programming and data analysis skills. The chosen scholar will follow in the footsteps of previous generations of scholars who have contributed to this project over the past 10 summers, many of whom have gone on to highly successful research careers that were kicked off by their project.
Research Environment: The student will work in tight collaboration with a young and highly dynamic research group, consisting of Professor Andrew Dzurak, a number of postdoctoral research fellows and around 8 other research students. High-speed computing facilities and advanced semiconductor device modeling packages will be made available. The scholar will also have their own work-station available within the Centre research offices. The research group deals with all the aspects of the construction of a quantum computer: design, simulation, nanofabrication, fast electrical measurement at ultra-low temperatures, and data analysis.
Novelty and Contribution: Our group has recently demonstrated the world’s first two qubit logic gate in silicon, based on "quantum dot" devices which are remarkably similar to existing silicon CMOS transistors. There are 2 options for the selected student to work on. First, the student can develop a new software tool (or driver) that will be able to assist in fine-tuning a multi-qubit device into its operation mode, and to readout the qubit states with high fidelity (or accuracy). Second, the student can work on simulating electron spin control pulse shape to provide better qubit control fidelity. Both options would be tested on our qubit devices currently under measurement at the low temperature dilution refrigerator.
Expected Outcomes: At the end of the Taste of Research period, in option 1, the student should have coded (in MATLAB) a simulator for a simple quantum dot system. This code will automate the fine-tuning process of our qubit operating point. In option 2, the student should have simulated various control pulse shape that can be tested in current qubit measurement. Beside this specific research outcome, the student will have been exposed to a wide range of advanced modeling, electronics, device engineering and physics concepts, and will have been involved in the daily challenges of an extraordinarily advanced research project. Our group has supported Taste of Research scholars for each of the past 10 summers, and in each case their positive experiences of the research here has inspired them to continue to an Honours thesis and PhD project with us.
Reference Material Links: Introductions to spin-based quantum bits, including recent media coverage of our work available [1,2,3]. Free-to-read publications that describe our silicon MOSFET quantum dots (paper 4 and 5) and recent success in observing and manipulating single spins in silicon (papers 6 and 7). A link to the MATLAB package is given in [8] and [9] is Prof. Dzurak’s group website.
[1]http://spectrum.ieee.org/computing/hardware/key-step-toward-a-silicon-quantum-computer
[2]http://www.abc.net.au/news/stories/2010/09/27/3022876.htm?section=justin
[3]http://www.scienceinpublic.com.au/media-releases/qc_nature#more-4054
[4]http://arxiv.org/abs/1003.2679
[5]http://arxiv.org/abs/0910.0576
[6]http://arxiv.org/abs/0904.0311
[7]http://arxiv.org/abs/1411.5760
[8]http://au.mathworks.com/
[9]http://blogs.unsw.edu.au/dzuraklab/
Other contacts:
PhD candidate, Mr. Wister Huang: wister.huang@unsw.edu.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Silicon-based Quantum Computing
Name of Supervisor: Scientia Professor Andrew Dzurak
Email of Supervisor: a.dzurak@unsw.edu.au
Name of Joint/Co-Supervisor: Dr Henry Yang
Email of Joint/Co-Supervisor: henry.yang@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): MEMS, Micro & Nano Technologies
School Research Area: Quantum Computing and Microelectronics
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Photovoltaic and Renewable Energy Engineering
Sciences – Maths, Physics, Chemistry
Abstract: The Centre for Quantum Computation and Communication Technology (see www.cqc2t.org) is focused on the fundamental physics and technology of fabricating a revolutionary silicon-based solid state quantum computer prototype. Quantum computers represent the next generation technology in computing and electronics. Through manipulation of quantum states, they offer parallel processing power and capacity in applications of commercial and national significance.

The simplest and most convenient implementation of a quantum bit is the spin of a single electron. UNSW hosts the world-leading group in the control and detection of single electrons and single spins in silicon. Our group has recently demonstrated the first ever single-shot measurement of a single spin in silicon [see Nature vol. 467, p. 687 (2010)] and has also developed a new type of MOSFET-based “quantum dot” in silicon in which individual spins can be confined and controlled.

This Taste of Research project will focus on the development of silicon quantum-dot qubits, using advanced modeling and measurement techniques available in our laboratory. The chosen Scholar will follow in the footsteps of previous generations of scholars who have contributed to this project over the past 7 summers, many of whom have gone on to highly successful research careers that were kicked off by their project.
Research Environment: The student will work in tight collaboration with a young and highly dynamic research group, led by Scientia Professor Andrew Dzurak, and comprising 4 postdoctoral research fellows, around 10 other research students, plus numerous expert technical staff who will train the Scholar. High-speed computing facilities and advanced semiconductor device modeling packages will be made available. The Scholar will also have their own work-station available within the Centre research offices. The research group deals with all the aspects of the construction of a quantum computer: design, simulation, nanofabrication, fast electrical measurement at ultra-low temperatures, and data analysis.
Novelty and Contribution: Our group has recently demonstrated the world's first two-qubit logic gate in silicon, which is a crucial step on the pathway to building commercially relevant quantum computers.
Our qubits are based on silicon MOSFET "quantum dots", in which electron spins can be added to the device in a controlled way. We use the spin of electrons confined in the SiMOS devices to encode the qubit information. The student’s contribution will be in designing, modeling and testing novel quantum dot structures, in order to determine the optimal design for our qubits. They will use advanced semiconductor device simulation software to predict the operation of the qubit devices, steer the design of new devices and actively participate in their testing and operation.
Expected Outcomes: At the end of the Taste of Research period, the student should have designed, modeled and possibly tested a complete silicon qubit device. Beside this specific research outcome, the student will have been exposed to a wide range of advanced modeling, electronics, device engineering and physics concepts, and will have been involved in the daily challenges of an extraordinarily advanced research project. Our group has supported Taste of Research scholars for 7 of the past 8 summers, and in each case their positive experiences of the research here has inspired them to continue on to a PhD project with us.
Reference Material Links: The links below include introductions to our research on spin-based quantum bits, including media coverage of our work [1, 2, 3]. Links are also included for publications that describe our success in manipulating single spins in silicon (paper 4), developing quantum dots (5 and 6), and our papers in Nature showing the first SiMOS qubits [7, 8].

[1] http://www.smh.com.au/technology/sci-tech/australian-researchers-make-quantum-computing-breakthrough-paving-way-for-worldfirst-chip-20151005-gk1bov.html

[2] http://spectrum.ieee.org/computing/hardware/key-step-toward-a-silicon-quantum-computer

[3]http://www.abc.net.au/news/stories/2010/09/27/3022876.htm?section=justin

[3] http://www.scienceinpublic.com.au/media-releases/qc_nature#more-4054

[4] http://arxiv.org/abs/1003.2679

[5] http://arxiv.org/abs/0910.0576

[6] http://arxiv.org/abs/0904.0311

[7] http://arxiv.org/abs/1407.1950

[8] http://arxiv.org/abs/1411.5760
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Friction-sensing-based force-controlled gripping
Name of Supervisor: Stephen Redmond
Email of Supervisor: s.redmond@unsw.edu.au
Name of Joint/Co-Supervisor: Heba Khamis
Email of Joint/Co-Supervisor: h.khamis@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Signal Processing & Control
School Research Area: Rehabilitation Engineering
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Mechanical & Manufacturing Engineering
Abstract: We have developed a tactile sensor which can sense incipient slip (that is, impending loss of grip) between an object and a robotic gripper. This sensors can also estimate the coefficient of friction in real-time when the incipient slip occurs. We have also manufactured a gripping actuator and control system which can apply a prescribed grip force to the object. The current focus of the project is to complete the instrumentation of the tactile sensor and marry it with the grip-force control system; the grip-force control system must also be extended to increase the grip force when there are also torques present at the gripping interface. The utility of the unified system will be tested by lifting objects with unknown weight and friction and rotating them in 3D using a six-axis Epson robot.
Research Environment: The work will be performed in the lower ground floor labs of the Samuels Building, part of the Graduate School of Biomedical Engineering. We have access to two six-axis robots, and PI hexapod, a precision xyz stage, a Stratasys Mojo 3D printer, and general electronics laboratory apparatus. The team currently consists of A/Prof Stephen Redmond, Dr Heba Khamis, Mr Han Wen (PhD student), Mr Wei Chen (PhD student).
Novelty and Contribution: This is ground-breaking research on the development and application of friction sensors. This area has largely been ignored by researchers in this field, who rather focus on increasing the resolution of force/pressure sensors, or building sensors that can sense shear forces; all of these sensors are of little use if no information is available about friction or grip security. Some have developed slip sensors, but usually just sense overt relative movement between gripper and object, which is a suboptimal approach, for a number of obvious reasons.
Expected Outcomes: Construct and instrument the tactile sensor (3D printing, silicone casting, electronics, mechanical assembly). Build a data acquisition system using an Arduino. Refined an Arduino-based grip-force control system. Experimental validation of tactile sensors performance using linear xyz stage. Experimental validation of control system using xyz stage. Advanced experimental validation of entire system using six-axis robot, rotating the grasped object in 3D.
Reference Material Links: Some early prototypes of one type of friction sensor we have developed. The second type has not been publicly disclosed yet:
http://ieeexplore.ieee.org/document/7319372/
https://link.springer.com/chapter/10.1007/978-3-319-42321-0_46
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Smartphone apps for clinical settings (cardiac rehab, gait analysis and mental health)
Name of Supervisor: Nigel Lovell
Email of Supervisor: n.lovell@unsw.edu.au
Name of Joint/Co-Supervisor: Michael del Rosario
Email of Joint/Co-Supervisor: m.delrosario@unsw.edu.au
School: Graduate School of Biomedical Engineering
Faculty Research Area (Theme): Programming Languages and Software Engineering
School Research Area: Rehabilitation Engineering
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Electrical Engineering & Telecommunications
Abstract: This work will be done in collaboration with colleagues from the Faculty of Medicine, Black Dog Institute, Prince of Wales Hospital and Sydney Children's Hospital to design apps for questionnaire delivery, social media usage and activity monitoring for assessing aspects of cardiac health, mental health and well-being.
Research Environment: Researchers and academics from GSBmE have been developing apps and telehealth technologies for over 15 years. Various tools, development platforms and database servers have been used but the current development will probably focus on emerging industry standards using React, node.js and Sequelize.
Novelty and Contribution: Very few apps have been developed that combine physiological measurements from sensors in the Smartphone and from Bluetooth connected measurement devices (such as blood pressure and weight scales) and then applied these measures to assess mental well-being.
Expected Outcomes: A functional app that will be tested in various clinical domains for assessing mental and physical well-being.
Reference Material Links: See supervisors for relevant materials.
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? Yes
If Yes, provide details*: Student may need to visit the hospital complex at Randwick to meet with clinicians to gather functional requirement data for design of their app.

[Back to Top]


Project Title: Advanced Signal Processing for GPS Anti-Spoofing
Name of Supervisor: Joon Wayn Cheong
Email of Supervisor: cjwayn@unsw.edu.au
Name of Joint/Co-Supervisor: Andrew Dempster
Email of Joint/Co-Supervisor: a.dempster@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Signal Processing & Control
School Research Area: Satellite Systems
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: GNSS Spoofing is a growing threat to the civil, commercial and aviation users which will compromise the on derived position information used for toll charging, navigation, law enforcement, surveillance, etc. To produce a spoof-resistant navigation solution, this project exploits recent research developments in other GNSS fields. The result of this research will produce an anti-spoofing detection and mitigation algorithm that is sensitive, robust, low-cost, and autonomous that is scalable to a dual-antenna system. Recent GNSS hardware receiver developments in our school have produced a dual-antenna time-synchronised receiver that will allow the validation of the anti-spoofing algorithm developed in this project.
Research Environment: Support for the research environment at UNSW is well defined within its current strategic plan. The research environment in ACSER is Australia’s strongest satellite navigation research group. Available to the researchers are two Spirent satellite simulators (GSS8000 and GSS6560), many receivers including three generations of our own Namuru receiver, many high-quality antennas, signal recording and replay equipment, and timer-counters for timing work which are all available to be used in this collaborative project.
Novelty and Contribution: This research is significant because it addresses the important problem of delivering a trustworthy position solution for civilian and commercial applications. The novelty of this research is to incorporate carrier phase information and dual-antenna into the existing P(Y)-based spoofing detection method. Thus, compared to the existing spoofing detection methods, the proposed methods are new in that:
(a) For a single-antenna receiver, it improves the performance and detection sensitivity of existing P(Y)-based spoofing detection by considering all the signals from multiple satellites via CD.
(b) For a dual-antenna receiver with known spatial separation, it incorporates carrier-phase information to achieve an infrastructure-less spoof detection and spoofing mitigation.
Expected Outcomes: Greatly enhance the spoofing detection sensitivity of the existing method.
Reduce the reference-rover architecture into a dual-antenna rover-only system so that the rover receiver can operate ubiquitously and autonomously. This discards the requirement of a trusted infrastructure to provide a reference P(Y)-like signal template.
Result in an anti-spoofing architecture that is low-cost and economically viable as it is simply a “software fix”.
Mitigate or suppress the threat of spoofing by utilising a dual-antenna system
Reference Material Links: http://acser.unsw.edu.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Cubesat Ground Station Command and Data Handling Optimisation
Name of Supervisor: Andrew Dempster
Email of Supervisor: cjwayn@unsw.edu.au
Name of Joint/Co-Supervisor: Andrew Dempster
Email of Joint/Co-Supervisor: a.dempster@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Embedded Systems and Communications
School Research Area: Satellite Systems
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: UNSW-EC0 and ISPIRE-2 are 2 cubesats designed and launched into space by UNSW in April 2017. In June 2017, the Cubesat team has established communications with the 2 cubesats. The student will develop novel solutions to collect, manage and analyse satellite telemetry and scientific data collected from the UNSW ground station and ground stations around the world. The student will also develop novel frameworks and human computer interface for commanding the satellites.
Research Environment: ACSER has exposure to 5 years of cubesat development, it has networks with various space industry entities including Space Industry Association Australia and collaboration with NASA. It also recently secured funding to form a Space Research Training Centre. ACSER is also affiliated with Sabre Astronautics which will is spearheading the efforts to revolutionise Ground Station management
Novelty and Contribution: Cooperative satellite data telemetry collection becomes complicated in scenarios when multiple ground stations are to cooperatively monitor and command multiple satellites. This research and development work will contribute to solving this problem by employing state of the art scripting and database technologies and moving away from old sequential and C/C++ techniques.
Expected Outcomes: A novel solution and framework for cooperative satellite data management that will prove to assist the data management for the two UNSW-developed satellites.This contribution will also help UNSW collaborate with other satellite teams to systematically monitor and command their satellites.
Reference Material Links: acser.unsw.edu.au//qb50
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Ocean Remote Sensing using GPS Reflectometry
Name of Supervisor: Dr Eamonn Glennon
Email of Supervisor: e.glennon@unsw.edu.au
Name of Joint/Co-Supervisor: Mr Ben Southwell
Email of Joint/Co-Supervisor: benjsouthwell@gmail.com
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Spatial Information Systems and Positioning
School Research Area: Satellite Systems
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Abstract: GPS satellites transmit signals that are not only used worldwide for navigation and timing, but also for remote sensing. GPS reflectometry (GPS-R) involves detecting and processing GPS signals that are reflected from the earth's surface to infer properties of the surface. Special GPS receivers capable of receiving reflected GPS signals, sometimes in conjunction with simultaneous reception of direct path GPS signals, are used to observe these reflected signals. Such receivers are flown on aircraft, UAVs or low-earth orbiting satellites.

In the case of GPS-R ocean remote sensing, sustained wind at approximately 10 m above the surface generates ocean waves and this affects the delay-Doppler Map measured by the GPS-R receiver. Models can be used to predict the DDM, while inverting these models allows the wind characteristics to be estimated. However, DDMs are often noisy, especially those measured from low earth orbit.

One area of investigation involves de-noising / filtering of the DDMs generated by space-borne GPS-R receivers with the aim of improving the resulting wind speed estimates. Another involves developing techniques to combine a sequence of DDMs or multiple DDMs and then integrating these over time.
Research Environment: UNSW via the SNAPLab and ACSER have been working on GPS for over 25 years. It has developed its own GPS receiver that has been flown on multiple space missions, it has two Spirent GPS simulators, a NordNav record and playback device and multiple FPGA based GPS receivers. Two of its staff (Dempster and Glennon) were involved in the development of Australia's very first GPS receiver in the early 90s.
Novelty and Contribution: The aim of this project is to improve the quality of the wave height, wind-speed and direction estimates obtained from GPS-R DDMs. This will be achieved by applying novel signal processing and filtering techniques to DDMs generated by spaceborne and airborne platforms.
Expected Outcomes: Using Matlab and Python, develop spatial and temporal filtering techniques to improve the DDMs generated by GPS-R receivers. DDMs generated by the TechDemoSat1 will be used as test-data sets for this work.
Reference Material Links: http://www.merrbys.co.uk/About%20Page.htm

Earth Observation with GNSS Reflections, REF: E-GEM-CSC-TEC-TNO01
http://www.e-gem.eu/file/uploads/eb5e98563d4c6c0e05d6f9cf77607faa.pdf
http://www.gfz-potsdam.de/en/section/space-geodetic-techniques/topics/gnss-reflectometry/
http://www.e-gem.eu

Gleason, S, "Remote Sensing of Ocean, Ice and Land Surfaces Using Bistatically Scattered GNSS Signals From Low Earth Orbit", PhD Thesis, December 2006
http://aoss-research.engin.umich.edu/missions/cygnss/reference/gnss-overview/Gleason_Thesis_GNSS.pdf
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Sensor Fusion for Positioning Integrity in Safety-Critical Vehicular Applications
Name of Supervisor: Joon Wayn Cheong
Email of Supervisor: cjwayn@unsw.edu.au
Name of Joint/Co-Supervisor: Andrew Dempster
Email of Joint/Co-Supervisor: a.dempster@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Spatial Information Systems and Positioning
School Research Area: Satellite Systems
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Depending on GPS/GNSS without the capability of validating its accuracy and validity can produce catastrophic consequences. Current techniques to validate GNSS/GPS-based positioning are limited and produces a large confidence interval which would be inadequate for vehicular applications where these intervals need to be within 10cm-1m based on Road Safety Australia's research. This project will adopt a multi-sensor approach and leverage on optimal filter techniques for network of vehicles to produce a much narrower confidence interval to satisfy such applications. The sensors involved will be high accuracy MEMS sensors and inter-vehicular ranging sensors.
Research Environment: Support for the research environment at UNSW is well defined within its current strategic plan. The research environment in ACSER is Australia’s strongest satellite navigation research group. Available to the researchers are two Spirent satellite simulators (GSS8000 and GSS6560), many receivers including three generations of our own Namuru receiver, many high-quality antennas, signal recording and replay equipment, and timer-counters for timing work which are all available to be used in this collaborative project.
The student will work on this project with DATA61 which has produced high-accuracy inter-vehicle ranging modules and collaborate with RMIT Melbourne which has prolonged expertise in the field of sensor fusion.
Novelty and Contribution: Vehicular applications of positioning technologies depending on GPS/GNSS are widespread in its use. In the impending development of self-driving vehicles, relying blindly on GPS/GNSS-baed positions can produce catastrophic consequences.
Expected Outcomes: This is currently an active area of research that intersects the sensor fusion community, GNSS community and the optimal filter community. Hence, this research work has the potential to produce high-ranking journal quality publications.
Reference Material Links: acser.unsw.edu.au
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Under Actuated Attitude Control for Cubesats
Name of Supervisor: Joon Wayn Cheong
Email of Supervisor: cjwayn@unsw.edu.au
Name of Joint/Co-Supervisor: Andrew Dempster
Email of Joint/Co-Supervisor: a.dempster@unsw.edu.au
School: School of Electrical Engineering and Telecommunications
For CSE and EET Projects: School Project
Faculty Research Area (Theme): Intelligent & Autonomous Systems
School Research Area: Satellite Systems
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Mechanical & Manufacturing Engineering
Sciences – Maths, Physics, Chemistry
Abstract: With the limited mass, power and volume available on cuebsat platforms eliminating the need for reaction wheels is desirable. However, this leaves only magnetorquers to perform the actuation resulting in an under-actuated cubesat and special considerations must be made when designing the controller. The purpose of this project is to design, implement and test a controller for underactuated (magnetorquer only) cubesats. A student with experience in control system design and rigid body kinematics would suit this project.
Research Environment: Working in a small team with PhD students and senior staff the studen will investigate controller designs for underactuated cubesats. Support for the research environment at UNSW is well defined within its current strategic plan. The research environment in ACSER is Australia’s strongest satellite navigation research group. Available to the researchers are two Spirent satellite simulators (GSS8000 and GSS6560), many receivers including three generations of our own Namuru receiver, many high-quality antennas, signal recording and replay equipment, and timer-counters for timing work which are all available to be used in this collaborative project.
Novelty and Contribution: This is a common problem in the small satellites (cubesat) community and the solution so far has been using a reaction wheel which consumes large amounts of power and require a large mass and volume allocation. I this problem can eb solved for magnetorquers, many missions requiring precise stately attitude/orientation control can do away without reaction wheels.
Expected Outcomes: The student will employ model predictive techniques to produce an a much more accurate controller by using only magnetorquers and without any reaction wheels.
Reference Material Links: acser.unsw.edu.au/qb50
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]


Project Title: Low noise aircraft, wind turbine and submarine technology
Name of Supervisor: Danielle Moreau
Email of Supervisor: d.moreau@unsw.edu.au
Name of Joint/Co-Supervisor: Con Doolan
Email of Joint/Co-Supervisor: c.doolan@unsw.edu.au
School: School of Mechanical and Manufacturing Engineering
Faculty Research Area (Theme): Fluid Dynamics and Thermal Engineering
School Research Area: Thermofluids
Applicable to other Engineering
schools/disciplines:
Computer Science & Engineering
Sciences – Maths, Physics, Chemistry
Abstract: Environmental noise pollution has been identified by the World Health Organisation as a global public health problem. The greatest contributor to increased environmental noise in most countries is strong growth in both land and air transportation. With the need for clean, low-carbon energy, rural communities are also experiencing increasing noise levels due to the addition of wind farms. A major source of the noise produced by a variety of engineering technologies including aircraft, wind turbines and submarines is flow interaction with an airfoil. A ToR student will perform fundamental wind tunnel experiments to understand the flow and noise generated by an airfoil.

The project will be excellent preparation for building a strong BE thesis project and/or a PhD.
Research Environment: This project will be conducted in the Aerospace Laboratory using our subsonic wind tunnel facilities and advanced flow and noise instrumentation. The ToR student will be joining a dynamic research group consisting of academics, postdocs, PhD students and technical staff. We provide an excellent research environment and support that will allow a ToR student to develop research skills in the field of aeroacoustics.
Novelty and Contribution: This research will provide new, fundamental knowledge concerning the processes controlling flow-induced airfoil noise production. By connecting the fine details of airfoil flow to noise generation, this project will assist future innovation in low noise technology.
Expected Outcomes: This research will result in a detailed experimental database of flow and far-field sound measurements. This work is expected to contribute to high quality international journal and conference publications.
Reference Material Links: For more information on Danielle Moreau: https://research.unsw.edu.au/people/dr-danielle-joy-moreau
Will the student visit the premises of an industry partner, or undertake any activity on premises external to UNSW? No

[Back to Top]