Telecommunications Research Projects

Wireless and Network Communications

The major research themes in wireless communications are: Channel coding and iterative receiver techniques; Space-time coding and MIMO techniques; Cooperative and cognitive radio communications; and wireless positioning and vehicular wireless ad hoc networks.  Major research themes in the data networks include: Quantum coding for communications; Wireless sensor networks; Security and trust; Greening the internet; and Networking for multimedia communications.  Recently, there has been significant cross-pollination between these traditionally distinct research areas.  One good example of this is the interaction between channel coding for physical communications and cooperative network coding; another is the interaction between channel coding, information theory and quantum communications.
Some of our ongoing research projects are as follows:

Multiuser MIMO Communications
Recent research shows that with multiple antennas placed at both the transmitter and receivers, referred to as multiple-input/multiple-output (MIMO) systems, wireless communication is able to increase the data rate significantly. This is a breakthrough in communications system design, since the multipath reflection in wireless channel, traditionally a pitfall of wireless communications, can be turned into a benefit, in increasing the wireless link capacity. Research is being conducted into novel transmission and multiple access signalling techniques with the aim of dramatically improving the reliability, throughput, and power efficiency of wireless downlink packet data services. This work is called the multiple-input/multiple-output (MIMO) spatial division multiple access (SDMA) technique and it will enable a breakthrough in multi-user multimedia services in the ICT industry sector. Potential applications of the project outcomes are in future 4G cellular mobile networks.

Cooperative Wireless Communications and Network Coding
The Wireless & Data Communications Research Group is working on developing novel transmission and receiving techniques with the aim of dramatically improving the reliability and throughput of wireless packet data services.   In particular, by exploring the space resource of multiple users, we develop cooperative multi-user communication techniques where multiple users or multiple base stations cooperate with each other to transmit their information. This can thoroughly exploit the space resource of multiple-users and user cooperation diversity in wireless networks to largely improve the reliability, and spectrum and power efficiency. With users cooperating with each other, we can significantly reduce the transmission power/energy, making the systems more “green”. Potential applications of the project outcomes are in future wireless systems, such as mobile broadband wireless access (MBWA 802.20 or WiMAX 802.16).

Cognitive radio networks
The Wireless & Data Communications Research Group aims to develop novel spectrum agile radio communication techniques which can opportunistically exploit the spectral resource of licensed systems and utilise the amount of unused spectrum in an intelligent way. The School’s current research outcomes in this area include robust cooperative spectrum sensing and whispering radio technique which enable to break the spectrum-availability bottleneck and significantly improve the spectrum utilisation.   The ultimate aim of the research is to dramatically improve the network’s spectrum efficiency, power efficiency and reliability, without interfering with other incumbent devices in the same frequency bands.

Improving Energy Efficiency of Internet Routers
The ICT sector consumed 156 GigaWatts, or about 8% of the world’s total electrical power consumption, in 2007, of which 14% is attributed to network equipment. The increasing amount of power consumed by Internet routers is becoming a serious concern for router manufacturers and Internet Service Providers (ISPs) alike. It is limiting the switching capacity router manufacturers can pack per unit space, and bloating operational expense for ISPs due to higher electricity bills and cooling costs. In this research project our aim is to develop innovative methods for energy reduction in Internet routers. We are aiming to develop new router architectures that employ more optics, optimise the use of components such as packet storage memories and interface speeds, and integrate emerging standards such as Energy Efficient Ethernet.

Networks for Environment Pollution Monitoring
Exposure to air pollution is known to increase the risk of cardiovascular and respiratory mortality, and exacerbate conditions such as asthma and chronic obstructive pulmonary disease (COPD). Current systems for air pollution monitoring have very poor spatial resolution, and do not reflect actual exposures experienced by individuals. In this project, we are building a system based on participatory sensor networks, whereby users with mobile phones contribute pollution data that is then collected centrally in real-time and displayed as a map. Alongside, we are developing tools that allow accurate estimation of personal exposure to air pollutants. Our research will help gain an understanding of urban air pollution distribution, as well as benefit individuals in understanding their personal health risk index.

Body-Area Networks for Healthcare Monitoring
This research develops energy-efficient communication protocols for body-wearable wireless sensor devices to be used in pervasive medical monitoring. Today’s healthcare systems are struggling to cope with the needs of an ageing population exhibiting an earlier onset of chronic conditions that need long-term monitoring. Wearable wireless sensors can relieve this pressure by providing intelligent, non-intrusive, continuous monitoring at dramatically reduced cost, with round-the-clock diagnostic and intervention capability. Our work in this area is developing the highly energy-efficient, light-weight, flexible, and robust communication protocols that are an integral part of such a system.

Reliable Communications for Vehicular Networks
In order to reduce traffic accidents, effective solutions are required. Vehicular Ad hoc Networks (VANETs), or Vehicular Communication (VC) systems, have potential to provide solutions and are likely to be ubiquitous in the not-too-distant future. In such systems, equipment exists on board the vehicles as well as in road-side infrastructure. DSRC is the method of communication for this network. DSRC is an alternative solution to a GPS-only-based solution, where the GPS in the vehicle may have access to an INS. By using DSRC, it can also use positioning information from other nearby cars and road-side infrastructure to enhance its own position accuracy and availability.

Quantum Communications
Quantum Communications is an emerging cross-disciplinary field of growing global significance.  Research into advanced quantum protocols is being pursued that will dictate the key operations of emerging quantum networks. Specifically, we are investigating the optimal quantum repeater protocols for a range of network architectures in which quantum information transfer through a multihop environment occurs. We will also determine near-optimal versions of our protocols that will give engineers the ability to trade off quantum complexity with communication throughput. New applications of quantum communications are also being researched that will bring enhanced security and communications advantages not possible in classical networks. Our work will result in new applications and services that will have a major impact on the ongoing global efforts to develop the quantum internet.

Photonics

The major themes in Photonics research are: Photonic fibre devices; Fibre based sensors; Planar photonic components and optical circuitry; and New photonic materials for sensing, display and difficult environments.
Some of our ongoing research projects are as follows:

Structural Health Monitoring of Civil Infrastructures
This work develops highly multiplexed fibre sensor systems for structural health monitoring and risk assessment of critical transport infrastructures. This work is in collaboration with Sydney University and industry partner RTA.

Fibre Ring Laser Based Intracavity Gas and Chemical Sensors
Fibre laser based sensor systems have great potential for high sensitivity gas and chemical detection. We develop fibre ring laser based sensor systems in collaboration with Tianjin University, China.

Fibre-based Photoreactor for Photoinduced Processes 
This research looks to address issues faced by solar induced photocatalysis through integrating improved particle and optic systems to increase photon efficiencies and harness a greater portion of solar/visible light. In collaboration with researchers in Chemical Engineering and Industry Chemistry, we develop an optical fibre photoreactor system that would effectively allow for improved utilisation of photons by the semiconductor surface.

Photosensitive Polymer Optical Fibres and Gratings
Polymer optical fibre Bragg gratings are useful for strain sensor applications for large dynamic range. We develop polymer optical fibres with higher photosensitivity and fabricating POF gratings for various industrial applications.

Porphyrin Technologies for Sensing in the Energy and Mining Industries
The project works on the next generation of “extreme” gratings and chemical sensors, both passive and active, primarily for applications in sensing within the petroleum and gas industries (but applicable across the mining industry). This is an international collaboration project in partnership with University of Sydney, Institute of Photonic High Technology, Germany and the Federal University of Technology, Brazil.

Diamond-based photonics
A/Prof Ladouceur directed the research effort at UNSW that led to the first scalable all-diamond integrated circuits using a combination of photolithography, reactive ion etching (RIE) and focused ion beam (FIB) techniques with important application in Quantum Key Distribution and Quantum Computing. This important work has been highlighted in New Scientist: “Diamond 'wires' – quantum computing's best friend”.

New photonics materials
A/Prof Ladouceur’s research effort is centred on the development of new photonics materials for display, biomedical and telecom applications. Of particular interest is the development of hydrogel-based electronics ink for conformal (flexible) displays, chiral (co)polymers for polarisation control and manipulation of polarisation in optical fibres, self-assembly of polymer for photoreceptors (pixels) definition in artificial retina and semiconducting polymers for artificial skin (tactile sensors).

Signal Processing

The major research themes in signal processing are: Image and video compression; Estimation and modeling of motion, depth and other multimedia features; Efficient and flexible multimedia communication; Paralinguistic speech processing to recover emotion, language and speaker information; Cochlear speech processing;  Sound field acquisition, compression and synthesis; Radar signal processing; and Signal processing for biological, biomedical and health monitoring applications.
Some of our ongoing research projects are as follows:

Image and Video Compression
The School has been conducting world-leading research in the area of compression of digital media.  One focus of this work is scalable compression technologies, which generated embedded bit-streams whose subsets can simultaneously target numerous resolutions, bit-rates and regions of interest. Another focus is efficient representation and estimation of structural information, including motion, depth and geometric structure. Work in this area has contributed and is continuing to contribute to several major international image and video compression standards.

Multimedia Communication
The School has also been very active in the development of efficient methods for image and video communication over networks.  One focus of this work is the development of highly efficient and computationally tractable hybrid-ARQ protection strategies for scalably compressed multi-media over lossy packet networks.  A second focus has been the development of algorithms and standards to facilitate efficient and flexible access to remotely located image and video.  Major outputs from our work include the core paradigm that underlies the JPIP standard (IS15444-9), a family of hybrid-ARQ protection algorithms collectively known as LR-PET (Limited Retransmission Priority Encoding Transmission), and commercial deployment of some of these research outcomes through the Kakadu software toolkit.

Biological Inspired Signal Processing
A collaboration between UNSW and a team from the Institute for Digital Communications at the University of Edinburgh (UoE) under the Biologically Inspired Signal Processing (BIAS), aims to develop novel algorithms for the study of non-stationary signals in general and bat echolocation calls in particular. Many of the engineering (and more specifically signal processing problems) we face have been addressed in nature, sometimes with astonishing degrees of specialisation and success. It is hoped that an improved understanding of natural systems would inspire novel technologies.

Signal Processing for Nuclear Magnetic Resonance Spectroscopy
This project is a collaborative effort with a group from the department of Biochemistry at the University of Cambridge and more recently with the Graduate School of Biomedical Engineering, and has the goal of investigating novel approaches for the processing of Nuclear Magnetic Resonance Spectroscopy data in order to enhance the detection and study of biologically active compounds such as metabolites and heparin. This can lead to the unmasking of low concentration metabolites in a biological sample thereby contributing to the study of disease, toxicity, gene expression as well as drug development.

Paralinguistic Speech Processing
Processing and recognition of the linguistic content of speech has been a major focus for speech processing research for some decades, however more recently attention is shifting towards non-linguistic speech information, such as speaker identity, emotion and cognitive load. The research effort within the school aims to characterise this information, towards improving recognition accuracies in a range of applications. Collaborators include the Institute for Infocomms Research (Singapore), the Australian National University, and National ICT Australia.

Genomic signal processing
This project concentrates on developing new period estimation techniques and significance measures for characterising structure within symbolic sequences such as DNA, in particular the nucleosome, whose function in evolutionary dynamics is a current area of major research interest in biology. Jointly with the John Curtin School of Medical Research at ANU, the new approaches are being evaluated on whole-genome data.

Signal processing methods for triaxial accelerometry
In this project, models and analysis methods are developed for automatically determining the type of terrain and gradient being traversed by a subject wearing a triaxial accelerometer. This information is critical to the accurate determination of the energy expenditure of the subject, which in turn has important applications in biomedical engineering and clinical medicine.