This page provides outline descriptions of honours and graduate (M.E./MPhil/PhD) research topics that I currently have available. If you are interested in one of these topics please contact me. I am willing to reserve a topic for a reasonable period while enrollment arrangements are made. Because enrollments do not always go ahead as planned, you should not regard a topic as unavailable simply because it is tentatively assigned.
Please note: As of 2011 I am reaching full capacity. I will not accept students who cannot show strong ability and affinity to a topic.
Energy scavenging is a popular concept for powering low power systems, for example sensing and monitoring functions in remotely mounted or inaccessible locations and in security applications (prox cards). Energy to power such systems can be supplied by a variety of sources. Many of these have the common feature that the available voltage can sometimes be very low, and needs to be transformed to a supply of typically more than 2V to run the function in question. Voltage multipliers in monolithic form have been around for many years. Dickson's classic paper (JSSC 1976) has shown the way for almost all subsequent designs and variants.
Silicon-on-Sapphire (SOS) CMOS technology offers various advantages over plain CMOS on silicon substrate. The differences give SOS advantages in a variety of niche applications such as RF switches. We believe certain of the advantages of SOS may provide for a more efficient dc-dc converter, especially when the power supply is very low, well below 1V.
A key component of the approach will be the development of an oscillator that starts reliably at as low a voltage as possible. Other challenges include design of a charge pump with very low parasitic capacitance and optimising threshold modulation through body effect.
This project is a collaboration between Silanna in Sydney and Bill Redman-White at Southhampton in the UK. The project aims to capitalise on Silanna's fab capability, Professor Redman-White's CMOS analog experience, and Professor Scott's oscillator experience. Familiarity with Microwave Office, analog design experience, and wafer-level test and measurement would all be advantages.
For decades, the microwave tool of choice has been the VNA (Vector Network Analyser). This device measures S-parameters (impedance, gain, etc.) with unprecendented precision from Hz to hundreds of GHz. This feat is achieved through the use of "vector correction", a calibration procedure that is able to cancel out systematic errors in the high-frequency measurements. The VNA operates with "small-signal" sine waves, and yields results as a function of frequency. S-parameters are defined only for linear systems, hence the VNA restriction to small signals.
Nonlinear measurements systems were developed throughout the 1990s. A Large-Signal Network Analyser (LSNA) was first available for purchase circa 2001, and the much more versatile and cheaper "Nonlinear VNA" was released in 2008 as the Agilent N5242A PNA-X. (Professor Scott was one of the designers of this instrument. See P. S. Blockley, D. Gunyan, and J. B. Scott, "Mixer-Based, Vector-Corrected, Vector Signal/Network Analyzer Offering 300kHz-20GHz Bandwidth and Traceable Phase Response," in Proceedings of the International Microwave Symposium, 12--16 June 2005.) This instrument measures "nonliner S-parameters" as a function of stimulus frequency, but can also operate as a precision time-domain analyser, although it measures frequency components sequentially using a mixer-based receiver and transforms the result using an FFT. The advanatage of this system is that vector correction produces a traceable, precise result, unlike an oscilloscope.
Recent developments in the definition and measurement of time-domain waveform characteristics reflect the desire to achieve precision time-domain measurements at frequencies previously impossible for oscilloscopes and other time-domain instruments. Meanwhile, real-time oscilloscopes (capable of measuring a single event) have appeared with increasing bandwidth. In 1980, the fastest ocsilloscope in existence could measure up to 1GHz with 4V peak-to-peak sensitivity, or 250MHz at millivolt sensitivity. By 2000, mV sensitivity and 8GHz was possible. Currently scope bandwidths have reached 16GHz, and this is expected to exceed 25GHz in the next generation.This project is to apply vector correction to a real-time ("single shot") oscilloscope by means of directional elements. The candidate will construct a Vector-Corrected, Real-Time Waveform Analyser version of an NVNA using a 4-channel oscilloscope. The aim is to prove the feasibility and implement the first ever vector-corrected, real-time, time-domain, traceable-response, microwave-speed instrument.
This topic will be available from the second quarter of 2011.
Applications where power and data are transmitted down the same wires have been discussed periodically in the literature for many years. Control and monitoring of power and lighting in homes to reduce power wastage and improve convenience is desirable, for instance, but has not become widespread. One implementation of control over power distribution, Digital Command Control or simply DCC, invented by Lenz Elektronik and later accepted as a standard for use in model railways, has been routinely and successfully used for decades. In principle, the use of such a system brings the opportunity for greatly-simplified wiring and straightforward control of multiple devices on the one data bus. Unfortunately these simplifications bring with them a daunting burden for non-technical users. The need to specify equipment, assign addresses, differentiate actuators, implement control, and debug minor problems represents a barrier to adoption.
A study has shown that in practice only technical users adopt DCC. The use of commercial controllers is reminiscent of programming a VCR: Unintuitive interfaces and the need to understand details of the workings counter the appeal of its advantages. The user must address too many options, remember too many details.
This project is about the design of a different DCC implementation, hardware and user interface. It aims to produce "iDCC", something that is to DCC what a Mac is to a PC, or TiVo to the VCR. The project will require analysis of user needs, perhaps with formally-documented formative feedback from focus groups, the development and implementation of a complete system vision. This vision should take the complete standard into account, implementing as much as possible in a way that obviates need for the user to understand technical detail by means of intelligent algorithm design coupled with appropriate hardware. The project will involve human factors analysis, creative design, original electronics and intelligent embedded coding.
The DCC system is a suitable vehicle for this study for several reasons. Standards are well-controlled, easy for engineers to understand, and publically readily available. Widespread standardisation has lead to the availability of cheap hardware and highly-compatible products from numerous manufacturers. A significant level of acceptance in the user community means that users and would-be users are available for formative and summative evaluation of user perception, needs, and desires. The DCC system is a microcosmic model of bus-control systems, offering the usual levels of implementation from the physical through protocol to application layers, and is thus suitable as a model of any conceivable system of the future. Additionally "playing with trains" can be fun and should lighten what might otherwise be a long and heavy project.