IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 45, NO. 1, FEBRUARY 2003
55
Analysis and Simulation of a 1–18-GHz Broadband
Double-Ridged Horn Antenna
Christian Bruns, Student Member, IEEE, Pascal Leuchtmann, Member, IEEE, and Ruediger Vahldieck, Fellow, IEEE
Abstract—A 1–18-GHz broadband double-ridged horn antenna
with coaxial input feed section is investigated. For the ridged horn
antenna it is found that the radiation pattern, contrary to common
belief, does not maintain a single main lobe in the direction of the
horn axis over the full frequency range. Instead, at frequencies
above 12 GHz, the main lobe in the radiation pattern starts to split
into four large side lobes pointing in off-axis directions with a dip
of up to 6 dB between them along the main axis. Although this type
of horn is the preferred test antenna, which is in common use for
over four decades, no explanation for this unwanted behavior was
found in the open literature. To investigate this phenomenon in detail, a method of moments approach has been adopted to simulate
the complete antenna system. The simulations are in good agreement with the measurements over the 1–18-GHz operational bandwidth and indicate that the use of this type of horn antenna in EMC
applications remains questionable.
Index Terms—Broadband ridged horn antenna, method of moments, radiation pattern deterioration.
I. INTRODUCTION
H
ORN antennas are widely used devices for transmission
and reception of electromagnetic waves in areas such
as standard measurement equipment, EMC testing, radar, and
communication systems. Generally, they are simple to build,
provide very good directional performance, and show excellent
peak power handling capability. To extend the maximum usable
to, e.g., 18:1,
bandwidth of horns from typically
ridges are introduced in the flared part of the antenna. This is
commonly done in waveguides to lower the cutoff frequency of
the fundamental mode and thus expand the single-mode range
before higher order modes occur [1]. The same effect is exploited in ridged horn antennas [2]–[5]. The main motivation to
mode)
ensure single-mode operation (e.g., of the pseudoover the largest possible bandwidth compared to standard horns
without ridges is twofold: First, since the design of this antenna
type dates back to the late 1950s, numerical field solvers were
unavailable at that time, and the “workhorse” to calculate and
design these antennas was the RF network theory, which is
based on single-mode propagation [6]; second, propagation of
higher order modes, if not properly taken into account, will
lead to undesired effects in the radiation characteristics of the
antenna. This is typically the case at frequencies higher than
GHz, for a 1–18-GHz horn, as shown in this paper.
Also more advanced methods such as the Chu-formulation or
Manuscript received December 4, 2001; revised June 28, 2002.
The authors are with the Laboratory for Electromagnetic Fields and
Microwave Electronics (IFH), Swiss Federal Institute of Technology (ETH),
Zurich 8092, Switzerland (e-mail: bruns@ifh.ee.ethz.ch).
Digital Object Identifier 10.1109/TEMC.2002.808022
the -field-model are not suitable for predicting accurately
enough, e.g., the current densities within the ridge horn antenna
or its radiation characteristics [7], [8]. These methods are to a
large extent semi-analytical and require significant simplifications of the antenna geometry before they can be applied (e.g.,
they are not capable of including the coaxial feed network).
Broadband ridged horns are standard components that can be
bought off the shelf from different antenna manufacturers. However, regardless of the manufacturer, all 1–18-GHz ridged horn
antennas investigated by the authors exhibited a similar deterioration of the radiation pattern over a large range of higher
frequencies: While sweeping the frequency, sudden significant
gain reductions on the center 0 axis as well as other strong pattern deformations were measured without any apparent reason.
These problems are not mentioned in any manufacturer’s data
sheet, instead the antennas are usually delivered with an approximate gain curve being the only specification. Deterioration of
the radiation pattern is of particular concern for those applications where one relies on a “well-behaved” pattern such as
for EMC measurements, primary reflector feeds or calibration
purposes.
In the past, several electromagnetic simulations of ridged
horn antennas were made, but not over such a broad operational
frequency range and also not including the coaxial feed, e.g.,
[9]–[15]. Restrictions concerning the geometrical complexity
were generally made due to limitations inherent in the applied
numerical technique or for the sake of the available computational power. For the reasons outlined above, the decision
was made to use a full-wave method of moments (MoM)
approach to carry out a complete broadband simulation of the
ridged-horn antenna including the coaxial feed [16].
II. HORN-ANTENNA MODEL
The ridged horn consists of a coaxial type input connector,
a cavity below the coaxial input section, two exponentially
plane flares
shaped ridges, two lower as well as upper
and two wedges (Fig. 1). The boxed dimensions of the horn
are 184 126 112 mm . The specially manufactured
connector used for this antenna is a high precision type to
prevent the excitation of higher order modes up to 20 GHz and
to support a maximum power level of 5 kW. The plane side
walls in our horn antenna were originally made out of solid
aluminum tubes but were replaced by thin copper straps, which
can be more easily removed for comparison purposes. They
serve as a means to control the width of the radiation pattern
at lower frequencies, but do not have any measurable effect at
frequencies above 4 GHz.
0018-9375/03$17.00 © 2003 IEEE
56
IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 45, NO. 1, FEBRUARY 2003
Fig. 1. Front view of the 1–18-GHz horn. Note the type N input connector,
the E plane copper straps, the two exponentially shaped ridges, the lower and
upper H plane flares, and the cavity.
Fig. 2. Simulation model of the broadband ridged horn antenna (9492 PEC
metallic triangles). See Fig. 3 for a detailed view of the coaxial feed section.
A. Modeling of the Broadband Ridged Horn Antenna
Modeling of the horn antenna was performed using the
commercial electromagnetic simulation software FEKO® and
the built-in Delaunay triangulation for the parts having straight
edges only [17]. The curved ridges of the antenna were modeled
using the two dimensional triangulation freeware program “Triangle,” which allows a very efficient control of key parameters
such as e.g., maximum triangle size and suppression of Steiner
points on the outer contour [18]. The shape of the edge contour
is defined with third order spline functions. Whenever possible,
electrical, magnetic and geometrical symmetries were used
throughout the modeling process to keep the computational
effort (i.e., CPU-time and memory requirements: RAM size
2 GB) within acceptable limits. The resulting model of the
Fig. 3. Coaxial input configuration in the model of the broadband ridged horn
antenna (cut view). At the left end of the coaxial feed line, the source elements
are situated radially between the inner and outer conductor.
ridged horn antenna consists of 9492 perfectly electrically
conducting (PEC) metallic triangles (Fig. 2).
Triangulation and wire segmentation refinement were adjusted so that the numerical conditions (such as area and length
discretization with respect to the frequency) imposed by the
utilized MoM approach are fulfilled for the maximum operating
frequency of 18 GHz and so that the problem can be solved
within a reasonable time. The triangle area was forced to be as
and the length of the line segments
small as
. To eliminate the generation of electromagnetic
was set to
artifacts resulting from different geometry discretizations,
the values for the triangulation and segmentation were kept
constant throughout the 1–18-GHz range. For this reason, the
surface discretization in the coaxial feed section will appear to
be rather coarse as compared to its actual size (Fig. 3).
Eight different feeding networks were closely investigated
with respect to their ability to excite a pure TEM wave on the
coaxial feed line without the existence of common modes and
current loops, which is generally a nontrivial problem for MoM
codes [19]. The feeding method yielding by far the best and
most reasonable results was the configuration consisting of
hollow, eight-cornered inner and outer conductors powered by
eight equally spaced cophasal voltage sources; they are situated
at the left end of the coaxial input line and oriented in a radial
direction (Fig. 3).
B. Numerical Procedure
The MoM technique is used to directly solve the electric-field
integral equation (EFIE) on the discretized surface of the horn
antenna. Point matching as well as the method of weighted
residuals is then applied, which yields the final current density
distribution on the metallic triangles and the wire segments of
the feed horn, respectively [16], [19]. This full-wave approach
requires approximately 1.6 GB of main memory for the matrix
inversion that is needed in the process of the numerical evaluation for 9492 PEC metallic triangles. Problems and erroneous
BRUNS et al.: ANALYSIS AND SIMULATION OF BROADBAND ANTENNA
Fig. 4.
57
Fig. 5. E plane radiation pattern of the horn antenna at 15 GHz (comparison
between measurement, standard antenna model simulation, and a refined
antenna model with gaps).
H plane radiation pattern of the horn antenna at 5 GHz.
field results can arise from the exclusive usage of the EFIE if
the discretized geometry is highly complex and the EFIE is
used at its mathematical limitations. An example are current
loops on very small parts of the structure, which do not make
sense from an electromagnetic point of view but still can exist
in the simulation model since they do not violate the conditions
imposed by the EFIE [19]. Therefore, all results were carefully
checked for erroneous current distributions. The solution to
this problem would be to use both the EFIE and the magnetic
field integral equation (MFIE). This, however, would lead to
a significantly longer computation time and was therefore not
implemented in the algorithm.
III. SIMULATION AND MEASUREMENT
Measurements of the feed horn radiation pattern and the absolute gain within the 1–18-GHz range were carried out in an
plane. The raanechoic chamber both in the - and in the
diation patterns were recorded with an angular resolution of 3
over a 180 spatial range and compared to the simulation results derived with the MoM approach. Due to the vast amount
of data obtained from the measurements and simulations, only
exemplary results exhibiting typical phenomena and problems
are presented in the following subsections. A distinction is made
between results for lower and higher frequencies as well as for
the standard ( -/ plane) and a 45 slanted antenna measurement position.
A. Results for Frequencies
12 GHz
Fig. 4 shows the radiation pattern derived at 5 GHz in the
plane of the broadband ridged feed horn antenna. With the exception of the deviations in the outermost angular range, the agreement between simulation and measurement is very good. The differences at the outer angles are a result of the inequality of the
space surrounding the feed horn in the simulation scenario and
the measurement setup (numerically infinitely extended space
versus finite anechoic chamber volume with weak scatterers).
Due to the maximum usable geometrical length inside the anechoic chamber, also the plane wave far field condition
(1)
wherein is the speed of light and the largest diagonal extension of the antenna under test, will be violated for frequencies
GHz which results in incorrect measurements especially
Fig. 6. H plane radiation pattern of the horn at 13 GHz (comparison between
measurement, standard-antenna model simulation, and a refined-antenna model
with gaps).
in the lower side lobe region [20]. Further differences between
the simulation and the measurement results particularly in the
side lobe range are due to limited discretization possibilities in
the feeding area resulting in local violations of the energy balance. After the evaluation of all simulated and measured radiation patterns it can be generally stated that the agreement is
very good both in the - and plane for all frequencies lower
or equal to approximately 12 GHz.
B. Results for Frequencies
12 GHz
For frequencies greater than 12 GHz however, the discrepancy between simulated and measured radiation patterns was
initially considerable. The off-the-shelf ridged horn antenna
was carefully examined to pinpoint possible features distinguishing the “real” prototype antenna from the numerical
simulation model. Since the ridged horn cannot be made from
one piece but is assembled from several subsections, it was
quickly found that the mechanical accuracy of the subsections
did not match standard Swiss mechanical precision. The
bottom of the cavity and also the flaps were not connected
mm to
properly, exhibiting a gap on the order of
the neighboring structures. These gaps have been included in
the refined antenna model (labeled as “Gap Simulation” in the
radiation patterns) and proved to have a significant influence
on the simulation results. This is illustrated in Figs. 5 and 6
(“Standard Simulation” refers to the initial “ideal” horn antenna
model without gaps). Now, the overall agreement between
58
IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 45, NO. 1, FEBRUARY 2003
Fig. 7. Three-dimensional logarithmic radiation pattern of the broadband
ridged horn at 10 GHz (qualitative representation of gain; maximum gain is in
broadside 0 direction).
simulation and measurements is considerably better and the
gap simulation also reproduces the deteriorated center axis
pattern in the measured radiation pattern correctly.
The best agreement between measurement and simulation
of the radiation pattern considering the aforementioned gaps is
achieved at around 15 GHz in the plane. It should be noted,
however, that these improvements cannot be simply transferred
to all other simulations. In some frequency ranges, the gaps have
a less pronounced effect and their influence on the - and
plane patterns can also be quite different. Furthermore, due to
the much larger conduction currents in the cavity bottom, the
gaps in the cavity edges do have a stronger effect on the radiation pattern in the simulation model than the ones between adjacent flare parts. The knowledge however that very small (as
compared to the rest of the geometrical structure) gaps and geometrical imperfections do have a significant impact on the radiation pattern, allows an optimization of the broadband ridged
horn both in the simulation and in the actual antenna. Still, some
differences in the side-lobe region between the simulated and
the measured radiation patterns are present for higher frequencies, while the agreement in the near vicinity of the center axis
is good over the whole operational bandwidth. This is also true
for the absolute maximum gain in the - and plane measurements as well as for the absolute broadside gain (the latter being
defined as the gain in the 0 center axis direction).
C. Off-Axis Side Lobes
As was mentioned in the introduction, all of the commercially
available 1–18-GHz ridged horn antennas investigated by the
authors displayed a significant pattern deterioration at higher
frequencies. This deterioration, which results in a sudden gain
reduction in the -/ plane measurements, can be nicely visualized in a three-dimensional (3-D) simulated radiation pattern
plot. For frequencies below 12 GHz, the radiation pattern still
looks as expected from a typical horn antenna, i.e., it has one
Fig. 8. Three-dimensional logarithmic radiation pattern of the broadband
ridged horn at 15 GHz (qualitative representation of gain; maximum gain is in
direction of the four side lobes).
Fig. 9. 45 azimuthally slanted plane radiation pattern of the horn antenna at
15 GHz (comparison between measurement and simulation).
dominant main lobe (Fig. 7). At 15 GHz however, Fig. 8 shows
a pattern with quite undesirable characteristics: The single main
lobe is now split into four large side lobes that “grow” equally
spaced around the 0 center axis while the main lobe appears to
be strongly indented. Since the radiation pattern is measured in
practice only in the -and plane, measurements will consequently just show a sudden gain drop in the graph for the broadside direction. The large side lobes however will not be detected
in a standard -/ plane testing procedure and therefore do not
appear on pattern measurement data sheets.
In order to confirm the existence of these side lobes, the measurement setup was changed and the horn antenna brought into a
45 azimuthally slanted position. The experimental verification
of the 3-D pattern depicted in Fig. 8 clearly shows the presence
of the side lobes in the measurement (Fig. 9). The presence of
these side lobes is particularly problematic for EMC applications such as testing in an anechoic chamber, which usually depend on a “well-defined” radiation pattern with only a single
main lobe [21].
BRUNS et al.: ANALYSIS AND SIMULATION OF BROADBAND ANTENNA
59
above 12 GHz; this in turn is due to a general flaw in the layout
of all similarly designed broadband ridged horn antennas.
Depending on the application (e.g., high vs. low power)
possible solutions to circumvent the undesirable performance
degradation are a complete redesign of the feeding section,
dielectric volume insets in the flared part of the antenna or a
dielectric lens on the aperture.
IV. CONCLUSION
Fig. 10. Magnitude of the time-averaged Poynting vector in the horn aperture
at 10 GHz (qualitative representation; maximum power density is in the dark
regions).
Fig. 11. Magnitude of the time-averaged Poynting vector in the horn aperture
at 15 GHz (qualitative representation; maximum power density is in the dark
regions).
TABLE I
COMPARISON OF 0 AXIS GAIN DETERIORATION BETWEEN DOUBLE-RIDGED
HORNS OF DIFFERENT MANUFACTURERS
To investigate this phenomenon in more detail, a closer
look was taken into the electromagnetic field configuration
within the flared section of the horn antenna. In the simulation,
one can clearly see that for frequencies reaching the critical
region around 12 GHz suddenly much more complicated field
configurations start to develop as compared to the fields at
lower frequencies. Also the power distribution over the horn
aperture exhibits a similar behavior: A comparison of the
contour plots of the time-averaged magnitude of the Poynting
vector at 10 GHz (Fig. 10) and 15 GHz (Fig. 11) indicates
that single-mode propagation is present for lower frequencies
mode), whereas multi-mode propagation sets
(pseudoin at approximately 12 GHz (the asymmetry in the plot is
due to the unsymmetric coaxial feeding). Radiation pattern
measurements of four double-ridged horns built by different
manufacturers confirm this observation (Table I). All of them
exhibit a comparable gain reduction in the 13–15-GHz range
and showed very similar -/ plane pattern deterioration. Our
simulations indicate that this is clearly related to the failure
to effectively suppress the propagation of higher order modes
We presented the electromagnetic simulation of a 1–18-GHz
broadband double-ridged horn antenna including a coaxial excitation. Our frequency-domain MoM simulations have been supported by measurements. This is the first time that such a complete antenna system was simulated in one step over the entire
frequency range. It was found that for satisfactory antenna performance, small geometric tolerances of the ridged waveguide
horn must be maintained and that the introduction of mechanical
imperfections into the simulation model significantly enhances
the agreement between simulations and measurements. Furthermore, our research indicates that all similarly designed broadband ridged horns exhibit the same performance degradation in
the upper frequency range, as they fail to effectively suppress
the propagation of higher order modes. Nevertheless, manufacturers of these antennas continue to advertise them as suited for
application over the full 1–18-GHz range.
ACKNOWLEDGMENT
The authors would like to thank Dr. U. Jakobus of Electromagnetic Software and Systems (EMSS) Ltd., South Africa, for
many helpful discussions and exchange of ideas for the implementation of the software package FEKO. Modeling of some
parts of the antenna was accomplished with the mesh generator
“Triangle” by Dr. J. Shewchuk of the University of California
at Berkeley. Thanks also to S. Wheeler and C. Maccio from the
mechanical workshop of IFH for fabricating the high precision
prototype horn antenna.
REFERENCES
[1] S. B. Cohn, “Properties of ridged wave guide,” Proc. IRE, vol. 35, pp.
783–788, Aug. 1947.
[2] A. Hizal and U. Kazak, “A broadband coaxial ridged horn antenna,”
in Proc. 19th Eur. Microwave Conf., Turnbridge Wells, U.K., 1989, pp.
247–252.
[3] K. L. Walton and V. C. Sundberg, “Broadband ridged horn design,” Microwave J., pp. 96–101, 1964.
[4] C. W. Gillard and R. E. Franks, “Frequency independent antennas – Several new and undeveloped ideas,” Microwave J., pp. 67–72, 1961.
[5] J. L. Kerr, “Short axial length broad-band horns,” IEEE Trans. Antennas
Propagat., vol. AP-21, pp. 710–714, Sept. 1973.
[6] S. Hopfer, “The design of ridged waveguides,” IRE Trans. Microwave
Theory Tech., vol. MIT-3, pp. 20–29, Oct. 1955.
[7] W. L. Barrow and L. J. Chu, “Theory of the electromagnetic horn,” Proc.
IRE, vol. 27, pp. 51–64, 1939.
[8] N. A. Adatia, A. W. Rudge, and C. Parini, “Mathematical modeling of
the radiation fields from primary feed antennas,” in Proc. IEEE Symp.
Electromagn. Compat., Sept. 1977, pp. 329–333.
[9] T. Wriedt, K.-H. Wolff, F. Arndt, and U. Tucholke, “Rigorous hybrid
field theoretic design of rectangular waveguide mode converters
including the horn transitions into half-space,” IEEE Trans. Antennas
Propagat., vol. 37, pp. 780–789, June 1989.
[10] C. Reig and E. Navarro, “FDTD analysis of E -sectoral horn antennas
for broadband applications,” IEEE Trans. Antennas Propagat., vol. 45,
pp. 1484–1487, Oct. 1997.
60
IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 45, NO. 1, FEBRUARY 2003
[11] F. Tristant et al., “F.D.T.D. method applied to the generation and propagation of short pulses,” in Ultra-Wideband, Short-Pulse Electromagnetics 3, C. Baum et al., Eds. New York: Plenum, 1997, pp. 279–286.
[12] F. Tristant, F. Torrès, P. Leveque, P. Jecko, C. Cruciani, and P. Noël,
“Transient study of radiated field of horn antennas,” in Proc. 10th Int.
Conf. Antennas and Propagation, Apr. 1997, pp. 1346–1349.
[13] R. Bunger, R. Beyer, and F. Arndt, “Rigorous combined mode-matching
integral equations analysis of horn antennas with arbitrary cross section,” IEEE Trans. Antennas Propagat., vol. 47, pp. 1641–1648, Nov.
1999.
[14] A. K. Bhattacharyya and G. Z. Rollins, “Accurate radiation and
impedance characteristics of horn antennas—A moment-method
model,” IEEE Trans. Antennas Propagat., vol. 44, pp. 523–531, Apr.
1996.
[15] R. Bunger, R. Beyer, and F. Arndt, “MLFMM analysis of ridged waveguide horns,” in Proc. ESA Millenium Conf. Antennas and Propagat.,
Davos, Switzerland, Apr. 2000.
[16] R. F. Harrington, Field Computation by Moment Methods. Piscataway,
NJ: IEEE Press, 1993.
[17] U. Jakobus et al.. (2002) FEKO—Field computations involving objects
of arbitrary shape, Stellenbosch, South Africa. [Online]. Available:
www.feko-usa.com
[18] J. R. Shewchuk. (2002) Triangle—A two-dimensional quality mesh
generator and delaunay triangulator. Carnegie Mellon University, Pittsburgh, PA. [Online]. Available: www.cs.cmu.edu/quake/triangle.html
[19] E. K. Miller, Ed., Computational Electromagnetics: Frequency-Domain
Method of Moments. Piscataway, NJ: IEEE Press, 1992.
[20] IEEE Standard Test Procedures for Antennas, IEEE Standards Board,
1990.
[21] M. Windler, Private communication, June 2001.
Christian Bruns (S’00) received the Dipl. Ing. degree in electrical engineering from the University of
Karlsruhe, Karlsruhe, Germany, in 2000. Currently
he is working toward the Ph.D. degree in electrical
engineering at the Laboratory of Electromagnetic
Fields and Microwave Electronics (IFH), Swiss
Federal Institute of Technology (ETH) Zurich,
Switzerland.
From August 1996 to May 1997, he was a Visiting
Student at the University of Massachusetts, Boston,
and the Massachusetts Institute of Technology, Cambridge, through a scholarship grant. His research interests include the numerical
analysis of general electromagnetic problems, antenna systems, reverberation
chambers, EMC, and mobile communications. In addition, he is interested in
economics, business, and administration.
Pascal Leuchtmann (M’98) received the diploma in
electrical engineering, and the Ph.D. degree from the
Swiss Federal Institute of Technology (ETH), Zurich,
Switzerland, in 1980, and 1987, respectively.
He was a Scientific Assistant in the electromagnetics group, working on theoretical electrical engineering, mainly general-purpose numerical calculation of electromagnetic fields. Currently, he is a Lecturer in electromagnetics and leads the research group
in research on EMC and antennas at the Laboratory of
Electromagnetic Fields and Microwave Electronics
(IFH), ETH.
Ruediger Vahldieck (M’85–SM’86–F’99) received
the Dipl. Ing. and the Dr. Ing. degrees in electrical engineering from the University of Bremen, Germany,
in 1980 and 1983, respectively.
From 1984 to 1986, he was a Research Associate
at the University of Ottawa, Ottawa, ON, Canada.
In 1986, he joined the Department of Electrical and
Computer Engineering, University of Victoria, Victoria, BC, Canada, where he became a Full Professor
in 1991. During Fall and Spring 1992/1993 he was
a Visiting Scientist at the “Ferdinand-Braun-Institut
für Hochfrequenztechnik”, Berlin, Germany. Since 1997, he is a Professor
of electromagnetic field theory at the Swiss Federal Institute of Technology,
Zurich, Switzerland. His research interests include numerical methods to
model electromagnetic fields in the general area of EMC and in particular, for
computer-aided design of microwave, millimeter wave, and opto-electronic
integrated circuits. Since 1981 he has more than 200 published technical papers
in books, journals, and conferences, mainly in the field of microwave CAD.
Prof. Vahldieck, together with three co-authors, received the outstanding publication award of the Institution of Electronic and Radio Engineers in 1983.
In 1996, he received the J.K. Mitra Award of the IETE for the best research
paper in 1995. He is the Past-President of the IEEE 2000 International Zurich
Seminar on Broadband Communications (IZS’2000) and President of the EMC
Congress 2003 in Zurich. He is Associate Editor of the IEEE MICROWAVE AND
WIRELESS COMPONENTS LETTERS and member of the Editorial Board of the
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES. Since 1992,
he has served on the Technical Program Committee of the IEEE International
Microwave Symposium, the MTT-S Technical Committee on Microwave Field
Theory, and in 1999, on the TPC of the European Microwave Conference. He is
the Chapter Chairman of the IEEE Swiss Joint Chapter on MTT, AP, and EMC.