Seminars from 2005
Woodpile EBG Antennas and Waveguide Components: Building Blocks for Future
Microwave, Millimeter-Wave and Optical Communications Systems
Speaker: Dr Andrew Weily
Date: Friday 14th October 2005 at 1.30pm
Venue: E6A202
Electromagnetic bandgap (EBG) materials, also known as photonic
crystals, are creating new possibilities for controlling and manipulating
the flow of electromagnetic waves. They are formed from dielectric and/or
metallic structures that are periodic in one or more dimensions. Within
the EBG material there is a range of frequencies where propagating modes
can be fully suppressed in one or more dimensions. This range of frequencies
is known as the EBG. Three-dimensional EBG materials have the potential
to provide greater control of the performance of antennas and waveguide
components due to their complete bandgap, which prevents modes from propagating
in all directions within the material.
This presentation will examine antennas and waveguide components based on
a three-dimensional EBG material known as the woodpile. Background information
on the woodpile structure and its potential applications will be discussed.
Results from both theoretical modelling and measurements for antennas and
waveguide prototypes operating at microwave frequencies will be presented.
Due to the scalability of electromagnetic theory, these results are also
valid in the millimeter-wave, terahertz and optical domains. These novel
devices may form the building blocks for future integrated systems based
entirely on three-dimensional EBG materials, with diverse applications that
range from millimeter-wave imaging to optical integrated circuits for telecommunications.
Finding MIMO
Speaker: A/Prof Linda M. Davis
Date: Friday 7th October 2005 at 1.30pm
Venue: E6A202
Foschini's work in the 1990s sparked a multi-antenna revolution in the wireless signal processing community: multiple-input multiple-output (MIMO) streaming promises huge capacity gains, and perhaps the solution for mobile high speed data in dense urban environments. In this talk, we will look at where the research has taken us, and some of the issues in making this technology a reality. Along the way, we will look at some recent work in MIMO CDMA, MIMO OFDM and look back at some old lessons on transmit, receive and multipath diversity. We'll talk about where the (re)search is focussed now, and what the MIMO reality might be for consumer electronics and defence applications in the future.
Space-Time Coding in Compact and Distributed Multiple Antenna Systems
Speaker: Jan Mietzner, Research Assistant, University of Kiel, GermanyDate: Wednesday August 31, 2005 at 9am
Venue: Division of ICS, Electronics Department. Building E7B Room: 100 (ground floor), See campus map for location.
The error performance of distributed space-time codes in a wireless system with multiple cooperating transmitters and a single receiver is analyzed. Due to the distributed nature of the system, the transmitted signals are subject to different average path losses. As the analysis shows, this effect leads to reduced diversity gains. This result is explained on the basis of an interesting duality between systems with distributed transmitters and systems with densely-packed transmitters: For the case of Rayleigh fading, it is proved that any system with distributed transmitters can be transformed into an equivalent co-located multiple-antenna system with correlated antennas. Specifically, the case of equal average path losses corresponds to an uncorrelated system. In order to quantify the performance loss resulting from distributed transmitters, a simple performance measure between zero and one is proposed. Finally, the duality concept is illustrated by means of a simple example.
Low-Density Parity-Check Codes: Factor Graphs, Iterative Decoding, EXIT Charts, and Information-Theoretic Bounds
Speaker: Dr. Ingmar Land, Aalborg University, DenmarkDate: Wednesday August 31, 2005 at 10am
Venue: Division of ICS, Electronics Department. Building E6A Room: 102 (ground floor), See campus map for location.
Low-Density Parity-Check (LDPC) Codes are among the most powerful channel codes for average to long code lengths. They were already invented in 1962 by Gallager, but they had to be re-invented in 1997 by MacKay and Neal before their capabilities were discovered: irregular LDPC codes can achieve the channel capacity.
LDPC codes are defined by parity-check matrices with only a small number of ones, and they are iteratively decoded on factor graphs defined by these parity-check matrices. Compared to Turbo codes, their decoding scheme is much simpler, and they are not patented - which makes them attractive for industry.
The talk starts with a basic introduction to LDPC codes and their iterative decoding scheme. The remaining part of the talk addresses the analysis and design of LDPC codes for large code lengths. Therefore, the decoder is interpreted as an iterative processor of mutual information and concepts of information theory are applied. The extrinsic information transfer (EXIT) chart method is first reviewed, and then it is generalized using the concept of bounding combined mutual information. This generalization leads to information-theoretic bounds on decoding thresholds, i.e., bounds on the minimal quality of a communication channel for error-free decoding.
Noise Dynamics in Distributed Bragg Reflector multiwavelength fibre laser
Speaker: Shilpa PradhanDate: Friday 19th August 2005, 1 - 2pm
Venue: E6A102
The design and performance of a single longitudinal-mode continuous-wave multiwavelength distributed Bragg reflector (DBR) fiber laser is described. Noise suppression in a multiwavelength DBR fiber laser has been reported using feedback control of the pump power. In-phase relaxation oscillations were reduced by 22dB, however, out-of-phase oscillations were not affected.
Slot Loaded Printed Antennas for Band-notched UWB Communication
Speaker: Tharaka Dissanayake, PhD Student, Macquarie University, ElectronicsDate: Tuesday 21th June 2005, 1 - 2pm
Antenna engineers have suggested band notched antennas to deal with possible interference between Wi-Fi and Ultra-wideband (UWB) pulse communication. Planar antennas loaded with slots have been proposed and even patented. Usually, slot length is approximated as half wavelength at the notch frequency. However, in case of printed antennas, other factors such as slot width, substrate thickness and permittivity can be critical in deciding the notch frequency. We have applied slot line theory to predict the notch frequency of a UWB printed monopole antenna with a loaded slot. It has been demonstrated that empirical formulas used in slot line theory can be effectively used in predicting resonant frequency of the loaded slot. In contrast to previous approximate methods, effects of substrate thickness, substrate permittivity and slot width on notch frequency can be examined with this new formulation. The method is mathematically simple and could assist in speeding up the process of band notched antenna design with full wave simulation. Furthermore, we have found the limitations of slot loading to create notches at higher frequencies; antennas with double slots (or coupled slots) will been presented and discussed.
Encircled Power Study of Focal Plane Field for Estimating Focal Plane Array Size
Speaker: Douglas Hayman, (Trevor Bird, Karu Esselle and Peter Hall).Date: Tuesday 14th June 2005, E6A202 1 - 2pm
The relationship between the size of a focal plane array in a prime focus dish and achievable scan angle is studied in order to find a starting point for designing such systems. A physical optics model is used to obtain focal plane fields and the radii in the focal plane required to encircle 3 dB and 1 dB below the incident power is calculated for a range of dish diameters and focal length over dish diameter ratios (F/D). The study found the focal plane fields are largely independent of dish size when the scan angle is expressed in beamwidths and gives guidance for the optimal F/D to achieve maximum field of view for a given focal plane array size.
Review of study-leave activities
Speaker: Associate Professor Graham TownDate: Tuesday 24th May 2005, 1 - 2pm
Graham will briefly present his observations of R&D in science and technology in Europe, and then describe some of the experimental and theoretical work undertaken whilst on study leave at the University of St Andrews, Aston University, and Dublin City University, where he worked on semiconducting polymer lasers, fabrication of microstructured polymer fibre, and integrated optical sensors, respectively.
Formal synthesis of circuits using Linear Matrix Inequalities
Speaker: Jeffrey Harrison, g2 Microsystems, formerly of Macquarie UniversityDate: Tuesday 3rd May 2005, 1 - 2pm
Given a set of devices (eg. a MOSFET, an inductor, a capacitor and a resistor), can we build a linear time-invariant circuit (eg. an amplifier) with given n-port (eg. 2-port scattering) parameters at a set of frequencies from those devices? Reflecting integrated circuit design considerations, multiple copies of each device are allowed and the inductors and capacitors can have finite Q. We derive a set of necessary conditions and show how these can be solved using fast convex optimisation algorithms developed since 1980. The conditions are a considerable generalisation of existing results such as that you cannot get gain out of a transistor above its maximum frequency of oscillation fmax. We also discuss whether these conditions are sufficient and hence whether an actual circuit can be synthesized.
A Dual-Band Planar Compact Artificial Magnetic Conductor
Speaker: Yading Li, PhD Student, Macquarie University, ElectronicsDate: Tuesday 12th April 2005, 1 - 2pm
Presently there is an increasing interest in artificial magnetic conductor surfaces as they demonstrate special properties, such as the in-phase reflection and high impedance surface conditions, which cannot be achieved by PEC surfaces. Photonic bandgap (PBG) structures and other periodic structures offer the possibility to design such artificial magnetic conductors. In this talk, a novel, dual-band, planar artificial magnetic conductor (AMC) structure is presented. For the same operating frequencies, the unit cell length is approximately one third of that of a previous dual-band AMC, which also did not require vias. This novel AMC surface also offers additional flexibility in controlling the operating frequencies and bandwidths.
Seminars - Second Semester 2004
Why use Polymers in Integrated Optics?
Speaker: David Johnson, PhD Student, Macquarie University, ElectronicsDate: Friday 5th November 2004, 1-2pm
Photonic devices for telecommunication systems are typically labour intensive to manufacture. Current market pressures require more performance at a lower cost from these devices. One path to this goal is via the technology of Integrated Optics. This presentation will provide an overview of integrated optics, examine some of the desirable characteristics of the materials used in the production of integrated optical devices, explore briefly the characteristics of currently used non-polymer materials and finally explore the pros and cons of using polymers.
A New Research Platform For Cochlear Implant Signal Processing
Speaker: Michael Goorevich, Cochlear, and MSc student, Macquarie University ElectronicsDate: Monday 27th September 2004, 1-2pm
The Australian manufactured Cochlear Implant has given the gift of hearing to more than 55,000 profoundly deaf people around the world. This number is expected to grow significantly over the next decade as improvements in the technology increase performance and convenience for the recipient. A key area in this regard is the audio signal processing techniques employed in a cochlear implant "sound processor", which is the portion of the system processing sound into cochlear stimuli. To offer improvements in signal processing typically requires many years research and development with recipients via bulky take home real-time "research processors" or non-real-time laboratory setups. In this Master's project, a new system has been developed which delivers an improved laboratory based platform offering real-time signal processing, while maintaining high reconfigurability and power for experimentation purposes. In this talk, the hardware and software platforms will be described, as well as general project progress and issues encountered along the way.
Soliton Data Storage Ring
Speaker: Yik Chong, MSc student, Macquarie University ElectronicsDate: Thursday 23rd September 2004, 2-3pm
The key element for future high speed TDM optical networks is the short term memory storage of data packets before rerouting, bit rate conversion. Optical fiber loops are ideal for short term storage of optical signals. Various approaches to enhance short term data storage in fiber loops have been demonstrated. Just like other long distance transmission system, such loops suffer general problems, eg, non-linear effects, deterioration of signals due to dispersion, noise added by amplification each round trip. Gain control is difficult: too much gain results in positive optical feedback and lasing while too little gain results in limited storage time. Linear and non-linear optical control techniques such as sliding frequency control and dispersion management would be used to address this issues.
A Wideband CMOS Integrated Receiver for Phased Array Applications
Speaker: Suzy Jackson, Australia Telescope National Facility and PhD student, Macquarie University ElectronicsDate: Friday 17th September 2004, 1-2pm
New silicon CMOS processes developed primarily for the burgeoning wireless networking market offer significant promise as a vehicle for the implementation of highly integrated receivers, especially at the lower end of the frequency range proposed for the Square Kilometre Array (SKA).
An RF-CMOS "Receiver-on-a-Chip" is being developed as part of an Australia Telescope program looking at technologies associated with the SKA. The receiver covers the frequency range 500 MHz to 1700 MHz, with instantaneous IF bandwidth of 500 MHz and, on simulation, yields an input noise temperature of <50 K at mid-band. The receiver will contain all active circuitry (LNA, bandpass filter, quadrature mixer, anti-aliasing filter, digitiser and serialiser) on one 0.18 µm RF-CMOS integrated circuit. This talk outlines receiver front-end development work undertaken to date, including design and simulation of an LNA using noise cancelling techniques to achieve a wideband input-power-match with little noise penalty.
SiGe integrated circuits for 60 GHz wireless applications
Speaker: James Howarth, Macquarie UniversityDate: Friday 3rd September, 2004 1-2pm
Recently the Japanese opened up 7 GHz of continuous and unlicensed bandwidth centred around 60 GHz, the largest of its kind. Similar bandwidths have also been opened up for unlicensed use by Australia and the US with Europe to likely follow soon.
Two important characteristics of this band are absorption by oxygen and narrow beam width. These inherent qualities lead to many varied uses such as Gigabit WPAN's , WLAN's and point-to-point links as well as possible use for vehicle collision avoidance radar. Such links would satisfy the increasing demand for network bandwidth driven by the ever-increasing number of users and their desire to use multimedia applications.
Traditionally integrated circuits targeted to this band have been designed using expensive semiconductor technologies such as GaAs and InP. Recent developments in SiGe however, could allow for a much more cost effective solution, something that is required for high penetration in the marketplace. My work looks at the design of RF circuits on SiGe for use in the 60 GHz band, with the aim of developing an entire transceiver on a single chip.
Performance Analysis of Web Server Load Balancing System
Speaker: Wilson Li, Macquarie UniversityDate: Friday 3 September 2004, 1-2pm
The talk will cover system performance analysis, markov chain, numerical analysis and discrete event simulation.
Title: Focal Plane Arrays for Radio Astronomy
Speaker: Doug Hayman, ICT Centre, CSIRODate: Friday August 20, 2004, 1-2pm
Focal plane arrays for reflector antennas have been in use in many applications for some time and in the last decade a number have been implemented in radio telescopes. Most of these have relatively widely spaced elements, usually with each element being treated as a single
beam in a multibeam configuration. In the radio astronomy community there is considerable attention being given to densely-sampled focal plane arrays with each beam being formed from the complex weighted sum of a number or the elements. The advantages of this approach is the flexibility to synthesise multiple beams, correct for aberrations due to mechanical misalignment or deflection, cancel interfering signals and optimise the overall G/T of the system.
This talk will cover some of the basics of focal plane arrays and current developments in the field. It will also discuss the direction of the PhD project.
Scattering by a Luneberg Lens partially covered by a metallic cap
Speaker: Nasiha Nikolic, ICT Centre, CSIRODate: Friday August 27, 2004, 1-2pm
A Luneberg lens (LL) has many applications due to its advantages: wide bandwidth, very wide field of view and capability of placing simultaneous beams that may be pointed in arbitrary directions.
In most configurations there are a number of feeds placed around the lens and beam scanning can be achieved by switching between the antenna elements. Our aim is to develop an integral equation method for the analysis of a lens with a number of feeds placed on its surface or between the dielectric layers.
A LL with integrated feeds was not considered in the past, and it was not possible to compare the results of this analysis with any other measured or predicted results. At this point, we were looking for a well known structure that we could use to verify our results. One suitable structure for comparison is a LL partially covered by a metallic cap. This structure is frequently used as a radar cross section (RCS) enhancer.
The Luneberg lens partially covered by a metallic cap was analysed in the past using representation of the fields in terms of spherical harmonics and using physical optics principles, ie replacing a conducting surface by an equivalent current representation. Although, good agreement between measured and predicted results was obtained by this method, its accuracy is limited to the validity of equivalent current approximations.
We developed a new method for the calculation of RCS of a Luneberg Lens where the surface current is represented as a linear combination of modes defined for conical waveguides. The results for RCS are compared with the results obtained in CST Microwave Studio and good agreement is obtained.