Keynote Lectures 2019
Karu Esselle
Professor Karu Esselle, IEEE ‘M (1992), SM (1996), F (2016), received BSc degree in electronic and telecommunication engineering with First Class Honours from the University of Moratuwa, Sri Lanka, and MASc and PhD degrees in electrical engineering from the University of Ottawa, Canada. He is a Professor of Electronic Engineering, Macquarie University, Sydney, Director of WiMed Research Centre and the Past Associate Dean – Higher Degree Research (HDR) of the Division of Information and Communication Sciences.
Karu was elevated to IEEE Fellow grade for his contributions to resonance-based antennas. Karu has authored approximately 600 research publications and his papers have been cited more than 7,500 times. Since 2002, his research team has been involved with research grants, contracts and PhD scholarships worth over 18 million dollars, including 15 Australian Research Council grants. His research has been supported by many national and international organisations including Australian Research Council, Intel, US Air Force, Cisco Systems, Hewlett-Packard, Australian Department of Defence, Australian Department of industry, and German and Indian governments. Karu’s awards include 2017 Excellence in Research Award from the Faculty of Science and Engineering, 2017 Engineering Excellence Award for Best Innovation, 2017 Certificate of Recognition from IEEE Region 10, 2016 and 2012 Engineering Excellence Awards for Best Published Paper from IESL NSW Chapter, 2011 Outstanding Branch Counsellor Award from IEEE headquarters (USA), 2009 Vice Chancellor’s Award for Excellence in Higher Degree Research Supervision and 2004 Innovation Award for best invention disclosure. |
ABSTRACT - Leaky-wave antennas: from niche applications to mass market
Since the discovery of efficient leaky-wave radiation from a slot in a wave guide by Oliner, leaky-wave antennas have attracted a lot of interest in applications that require beam scanning. Printed planar configurations of LWAs have become very popular, due to low cost. Half-width LWAs based on microstrip lines and substrate-integrated-wave guides have provided an additional advantage of narrow footprint. Professor Christophe Caloz (APS Distinguished Lecturer 2014-16) has brilliantly summarised in his distinguished lectures and illustrated how the problem of massive drop of antenna efficiency, when the beam is attempted to steer in the broadside direction, has been solved using Composite Right/Left-Handed (CRLH) structures and other methods.
After briefly reviewing such crucial historical milestones in LWAs, this presentation will focus on recent developments in LWA antenna research and practical outcomes, including some that have the potential to further extend applications of LWAs from current niche scanning applications to mass communication applications such as wireless local area networks and emerging 5G mobile communications.
One of them is fixed-frequency beam steering using only two values of bias voltages, for applications where sweeping the operating frequency is not possible [1]. Several methods of LWA fixed-frequency beam steering has been demonstrated, including one recently developed by the speaker’s team that requires only two bias voltage values to steer the beam. This is very promising for millimetre-wave communication systems such as Wi-Gig and potential millimetre-wave modes of 5G.
The principle underlying these LWAs is formation of a multistate radiating structure by cascading several binary reconfigurable unit cells. Thus, the basic building block of the antenna is a reconfigurable binary unit cell, switchable between two states. A macro cell is created by combining several reconfigurable unit cells and the periodic LWA is formed by cascading identical macro cells. Antenna beam is digitally steered in small steps by switching to different macro-cell states. Microwave prototypes based on this concept have demonstrated excellent beam steering over 30 degrees with negligible gain variation (of about 1 dB) and good input matching. As all switches in the antenna are binary, only two bias voltage values are required for beam steering, and the antenna sub-system can be controlled easily using digital electronics.
Other recent developments presented in the lecture include (i) steering two side beams simultaneously by sweeping the operating frequency, using the second higher order mode of a microstrip [2], (ii) dual-band beam scanning by frequency sweeping, with one beam scanning forward directions and the other one scanning backward directions [3], (iii) arrays of leakywave antennas with a combined beam in broadside direction [4], and (iv) tri-band leaky wave antennas.
At the end, selected topics suitable to future research in this area will be discussed.
After briefly reviewing such crucial historical milestones in LWAs, this presentation will focus on recent developments in LWA antenna research and practical outcomes, including some that have the potential to further extend applications of LWAs from current niche scanning applications to mass communication applications such as wireless local area networks and emerging 5G mobile communications.
One of them is fixed-frequency beam steering using only two values of bias voltages, for applications where sweeping the operating frequency is not possible [1]. Several methods of LWA fixed-frequency beam steering has been demonstrated, including one recently developed by the speaker’s team that requires only two bias voltage values to steer the beam. This is very promising for millimetre-wave communication systems such as Wi-Gig and potential millimetre-wave modes of 5G.
The principle underlying these LWAs is formation of a multistate radiating structure by cascading several binary reconfigurable unit cells. Thus, the basic building block of the antenna is a reconfigurable binary unit cell, switchable between two states. A macro cell is created by combining several reconfigurable unit cells and the periodic LWA is formed by cascading identical macro cells. Antenna beam is digitally steered in small steps by switching to different macro-cell states. Microwave prototypes based on this concept have demonstrated excellent beam steering over 30 degrees with negligible gain variation (of about 1 dB) and good input matching. As all switches in the antenna are binary, only two bias voltage values are required for beam steering, and the antenna sub-system can be controlled easily using digital electronics.
Other recent developments presented in the lecture include (i) steering two side beams simultaneously by sweeping the operating frequency, using the second higher order mode of a microstrip [2], (ii) dual-band beam scanning by frequency sweeping, with one beam scanning forward directions and the other one scanning backward directions [3], (iii) arrays of leakywave antennas with a combined beam in broadside direction [4], and (iv) tri-band leaky wave antennas.
At the end, selected topics suitable to future research in this area will be discussed.
Boris TomasicBoris Tomasic was born in Rijeka, Croatia, on August 8, 1945. He received the Dipl. Ing. degree in electrical engineering from Fakulteta za Electrotehniko, Univerza v Ljubljani, Slovenia, in 1971 and the M.S. and Ph.D. degrees in electrical engineering from Polytechnic Institute of New York, Brooklyn, NY, in 1975 and 1981, respectively.
From 1975 to 1980, he was a Research Assistant at Polytechnic Institute of New York, performing research on conformal and planar array antennas. Since 1981 he has been with the Air Force Research Laboratory (AFRL), Sensors Directorate, first at Hanscom AFB, MA and presently at Wright-Patterson AFB OH. From 2007 to 2011 he was the Technical Advisor in the Antenna Technology Branch wich has the responsibility for developing the technology base for antennas within the Air Force. Dr. Tomasic is responsible for planning and conducting research and development studies in the field of electromagnetic radiation, propagation and diffraction. His research interest is in the areas of antenna theory, planar and conformal phased arrays for airborne and space applications, mm-wave electronically scanned array antennas, digital beamforming, and metamaterials. From 1981 to 1990, he was adjunct faculty member at the University of Massachusetts teaching graduate courses in electromagnetics and antenna theory. Dr. Tomasic is a member of Sigma Xi, URSI Commission B, and senior member of IEEE. From 1986 to 1989 and from 1997 to 1999 he served as associate editor for IEEE Transactions on Antennas and Propagation. He is an Air Force Research Laboratory Fellow and recipient of the 57th Annual Department of Defense Distinguished Civilian Service Award. |
ABSTRACT - Self-Calibration and Mutual Coupling Compensation in Digital Beamforming Arrays - An Example: Finite Array of Parallel-Plate-Guide Fed Slits on a Ground Plane
Electronically Scanned Arrays (ESA) with Digital Beam-forming (DBF) are becoming the key component in com¬mercial and military advanced radar and comm systems. As known, DBF is a powerful technique used to enhance the antenna performance through multiple beams, adaptive pattern control, element pattern correction, self-calibration, and flexible time management concepts.
To fully exploit these advanced features and concepts that DBF arrays offer, it is crucial and necessary to perform the array calibration that works in real time and in the operational environment. Presently, the mutual coupling and digital errors in DBF arrays are combined in standard array calibration schemes performed at a single scan angle, typically broadside, and one frequency in anechoic chamber. The calibration errors increase at other scan angles. In addition, the method is not feasible in the field, i.e., in operational contested environ¬ments due to absence of the reference source, uncooperative transmitters i.e., interference, and unknown scattering from the environment.
Thus, there is a great need for the self-calibration and mutual coupling compensation in DBF arrays in real time while operating in contested environments. The potential ben¬efits are enormous: improved array performance with ultra-low sidelobes while no additional hardware or external reference sources are needed.
In the proposed method we divided the array architecture into two sectors: digital and analog. The digital sector is scan independent, whereas the analog sector is scan dependent. In the former case, the errors in digital channels, causing channel imbalances, will be compensated by injecting pilot/test signals at the element test ports to equalize channel imbal¬ances. While in the latter case, the analog, scan dependent errors are primarily due to mutual coupling causing different embedded element patterns or array edge effects, which in turn degrade the array patterns. The mutual coupling effects will be compensated for all scan angles and frequencies and should work in both, Tx and Rx modes of operation. With advancements of digital beamforming technologies, such as high speed (>60 Gsps) analog-to-digital (ADC) and digital¬to-analog (DAC) converters, the self-calibration and mutual coupling compensation including beamsteering, adaptive pat¬tern control for supression of jamers, health and status check, etc. can be executed simultaneously in real time. Thus the DBF is becoming the key enabler in high-performance multi-beam, multi-function ESAs.
In this work, we focus on mutual coupling compensation in DBF arrays. The technique we use is similar to [1] where embedded element patterns were determined from isolated
element pattern perturbed by the array elements via scattering coefficients. In contrast to [1], the present study directly calculates scattering matrix (which can also be measured experimentally) instead of calculating them from measured embedded element patterns. As in [1], the compensation for the mutual coupling can be accomplished by simply multiplying the received signals v by the inverse coupling matrix C^-1, i.e., vd~=C^-1*v where vd represents the unperturbed desired signals. The mutual coupling matrix is directly related with the array scattering matrix by C = I + S.
As discussed in [2], we demonstrated the method on a simple 1D array of parallel-plate-guide-fed slits on a ground plane. We selected this geometry because it can be analyzed analytically, and all errors due to approximations in the method, as well as numerical errors can be accurately accounted for. During the presentation analytical expressions for embedded element patterns of an infinite and finite arrays, as well as pattern of an isolated element will be discussed and compared. The analytical results are also compared with simulated results using COMSOL. Respective array radiation patterns were calculated for two cases: with and without mutual coupling. The metric for success was to compare the patterns of an array with isolated elements, which means with no mutual coupling (S=0) with an array of elements with (actual) embedded element patterns, i.e., edge effects.
As will be shown at the presentation, at least in the dominant mode approximation for the field in the slots, the coupling compensation was perfect.
To fully exploit these advanced features and concepts that DBF arrays offer, it is crucial and necessary to perform the array calibration that works in real time and in the operational environment. Presently, the mutual coupling and digital errors in DBF arrays are combined in standard array calibration schemes performed at a single scan angle, typically broadside, and one frequency in anechoic chamber. The calibration errors increase at other scan angles. In addition, the method is not feasible in the field, i.e., in operational contested environ¬ments due to absence of the reference source, uncooperative transmitters i.e., interference, and unknown scattering from the environment.
Thus, there is a great need for the self-calibration and mutual coupling compensation in DBF arrays in real time while operating in contested environments. The potential ben¬efits are enormous: improved array performance with ultra-low sidelobes while no additional hardware or external reference sources are needed.
In the proposed method we divided the array architecture into two sectors: digital and analog. The digital sector is scan independent, whereas the analog sector is scan dependent. In the former case, the errors in digital channels, causing channel imbalances, will be compensated by injecting pilot/test signals at the element test ports to equalize channel imbal¬ances. While in the latter case, the analog, scan dependent errors are primarily due to mutual coupling causing different embedded element patterns or array edge effects, which in turn degrade the array patterns. The mutual coupling effects will be compensated for all scan angles and frequencies and should work in both, Tx and Rx modes of operation. With advancements of digital beamforming technologies, such as high speed (>60 Gsps) analog-to-digital (ADC) and digital¬to-analog (DAC) converters, the self-calibration and mutual coupling compensation including beamsteering, adaptive pat¬tern control for supression of jamers, health and status check, etc. can be executed simultaneously in real time. Thus the DBF is becoming the key enabler in high-performance multi-beam, multi-function ESAs.
In this work, we focus on mutual coupling compensation in DBF arrays. The technique we use is similar to [1] where embedded element patterns were determined from isolated
element pattern perturbed by the array elements via scattering coefficients. In contrast to [1], the present study directly calculates scattering matrix (which can also be measured experimentally) instead of calculating them from measured embedded element patterns. As in [1], the compensation for the mutual coupling can be accomplished by simply multiplying the received signals v by the inverse coupling matrix C^-1, i.e., vd~=C^-1*v where vd represents the unperturbed desired signals. The mutual coupling matrix is directly related with the array scattering matrix by C = I + S.
As discussed in [2], we demonstrated the method on a simple 1D array of parallel-plate-guide-fed slits on a ground plane. We selected this geometry because it can be analyzed analytically, and all errors due to approximations in the method, as well as numerical errors can be accurately accounted for. During the presentation analytical expressions for embedded element patterns of an infinite and finite arrays, as well as pattern of an isolated element will be discussed and compared. The analytical results are also compared with simulated results using COMSOL. Respective array radiation patterns were calculated for two cases: with and without mutual coupling. The metric for success was to compare the patterns of an array with isolated elements, which means with no mutual coupling (S=0) with an array of elements with (actual) embedded element patterns, i.e., edge effects.
As will be shown at the presentation, at least in the dominant mode approximation for the field in the slots, the coupling compensation was perfect.