Leica AR25

White Paper
 February 2009

Lennon Bedford, Neil Brown, Justin Walford

Leica Geosystems AG
Heerbrugg, Switzerland

2
Biography as it defines the measurement reference point. In
Lennon Bedford graduated from the University of order to achieve the best performance there are
Otago in 2003 with a Bachelor of Surveying (Hons). many aspects to the antenna that must be consid-
He is currently an Application Engineer for GNSS Net- ered, including:
works and Reference Stations for Leica Geosystems
based in Heerbrugg, Switzerland.  Low elevation tracking
 Phase centre stability
Neil Brown holds a PhD in satellite geodesy from the  Multipath mitigation
University of Melbourne and has been involved with  Out of band rejection
GNSS research since 2000. He is currently Senior  Front to back ratio
Product Manager for GNSS Networks and Reference  Gain pattern
Stations for Leica Geosystems, based in Heerbrugg,
Switzerland. Reference station operators are generally reluctant to
change antennas because they are so important for
Justin Walford holds an M.Sc.E in Survey Engineering the site position and accuracy. However, there are
from the University of New Brunswick, and has been many new signals that are now available or planned
involved in GNSS applications and research since as part of modernized GPS, modernized GLONASS,
1990. He is currently Product Manager for GNSS Net- Galileo, Compass, QZSS and other satellite navigation
works and Reference Station Hardware for Leica or augmentation systems. These space segment im-
Geosystems, based in Heerbrugg, Switzerland. provements include signals transmitted on additional
frequencies to the GPS L1 and L2 and GLONASS L1
and L2 that are commonly in use today. Antenna
Abstract changes will be required to provide “all in view” track-
This paper introduces the AR25, a new multi constel-
ing. Most notable are GPS L5, Galileo E1, E2, E5a,
lation Global Navigation Satellite System (GNSS)
E5b, E6 and Compass B1, B2, B3. Hence it is also
choke ring antenna for precise geodetic applications.
important, aside from the factors listed above, that
This revolutionary new ‘3D’ choke ring design allows
an antenna supports these new signals to avoid the
better low elevation satellite tracking while maintain-
need to change the antenna again in a few years
ing the renowned performance characteristics of the
time.
traditional choke ring antenna such as smooth ampli-
tude and phase pattern, effective multipath rejec-
From an antenna design point of view however, wid-
tions and phase centre stability. The AR25 contains a
ening an antenna to track this range of frequencies
new ultra wideband Dorne-Margolin element to allow
creates many challenges to optimize the above char-
for superior reception of all existing and planned
acteristics for each frequency. The Galileo E6 and
GNSS signals, providing users with improved position-
Compass B3 frequencies are the most difficult be-
ing precision and reliability.
cause of their proximity to frequencies used by air
traffic control.
Introduction
The well known Jet Propulsion Laboratory (JPL) de- In this paper, the performance of the new wideband
signed choke ring antenna with a Dorne-Margolin AR25 choke ring is compared to the AT504GG, an
vertical dipole has been widely accepted within the existing high-end choke ring antenna based on the
reference station community. Many choke rings from original design from JPL. The AR25 uses an innovative
various manufacturers have been based on this de- 3D choke ring design in which the rings are at differ-
sign and used within the IGS and other reference ent heights and contain slots to allow dissipation of
station networks. However, antenna theory has unwanted Radio Frequency (RF) energy. This new
evolved since this antenna was made allowing for design helps to improve gain at the horizon while
innovative choke ring designs that provide better all- maintaining stable phase centre and pattern symme-
round performance. Reference station operators try for amplitude, phase and group delay. This allows
demand the highest performance antennas to allow for better reception and tracking of low elevation
the most accurate determination of site positions satellites, improved multipath mitigation and out of
and velocities and high quality Real Time Kinematic band rejection.
(RTK) corrections. The antenna is arguably the most
important part of any reference station infrastructure

3
In order to compare the antennas, various real world Table 1: Existing and proposed GNSS signals [MHz]
System L1/E1/E2/B1 L2/B2 L5/E5 E6/B3
and laboratory tests were conducted to evaluate key
performance criteria for each antenna. Anechoic
GPS 1575.42 1227.6 1176.45
chambers tests were used to assess the general an-
GLONASS 1598.063 - 1242.938 -
tenna design and expected performance for the fu-
1605.375 * 1248.625 *
ture signals. Empirical tests using the current GPS and
Galileo 1575.42 1176.45, 1278.75
GLONASS constellations were used to relate the
1207.14,
theoretical characteristics to real world performance.
1191.795
Compass 1561.098, 1207.14 1176.45 1268.52
New Signals 1575.42,
Future proofing the AR25 for the planned GNSS fre-
1589.742
quencies detailed in Table 1 brings many benefits to
SBAS 1575.42
reference station and network users. A nominal mod-
OmniSTAR 1525 - 1560
ernized GPS, modernized GLONASS and Galileo con-
and CDGPS
stellation will comprise 78 satellites. This level of
k = -7,6
coverage will bring new levels of:

 Satellite availability (allowing improved positioning

New Innovative 3D Choke Ring
in difficult environments such as urban canyons) Design
 Geometry (i.e. low GDOP, PDOP etc.) A typical choke ring antenna consists of several con-
 Productivity (reduced time to fix) centric ring structures that surround the central an-
 Reliability (improved ambiguity resolution, espe- tenna element (Figure 1). The choke rings, which are
cially in difficult environments) usually slightly more than one quarter of a GPS L2
 Redundancy (better ability of the system to detect wavelength deep, are designed to eliminate reflected
problems) signals and prevent the propagation of surface waves
 Precision (more precise signals, better modeling) near the antenna (Kunysz, 2001).

Increased satellite availability leads to improved ge-

ometry and redundancy of observations. This in turn
improves reliability and precision of position, impor-
tant in a host of applications utilizing GNSS signals.
The benefits of combined GNSS technologies for RTK
applications are examined in detail by Takac and Wal-
ford (2006).

The new AR25 3D choke ring design has been opti-

mized for maximum compatibility with the antenna
element, for the highest tracking performance.

Figure 1: The AT504GG – A traditional '2D' choke ring

antenna

The choke ring antenna has become the industry

‘standard’ for high end permanent reference stations
due to its proven phase center stability, smooth am-
plitude and phase pattern and low susceptibility to
multipath. The choke ring antenna is used as a
benchmark for antenna qualification as an IGS station
(IGS, 2007).

A significant weakness of the traditional choke ring

antenna is its poor reception and tracking of satel-

4
lites near the horizon. Signals from low elevation
satellites are very important for many applications of
GNSS because they help to de-correlate station
height and troposphere parameters. One of the main
design features of the 3D design of the AR25 choke
ring antenna (Figure 2) is the improved low elevation Figure 3: Cross-section of AR25 showing the groove
tracking. depths

Additionally the AR25 uses a new ultra wideband

Dorne-Margolin element. The new element is specially
constructed and tested to ensure consistent per-
formance across all bands.

Choke ring antennas are the preferred choice for

reference station installations partly because of their
durable construction. The AR25’s robust construction
ensures that the antenna will pass the test of time in
the harshest of environments. An optional weather
proof radome is also available.

Testing
The Anechoic chamber tests were conducted at the
David Florida Labs of Canadian Space Agency located
in Ottawa (Kanata) to determine the radiation pat-
tern and phase center offset and variation in benign
conditions (free of multipath). Data was collected at
23 GNSS discrete frequencies from 1175.3MHz to
1610 MHz. Spatially, data was sampled at 3 deg.
intervals in both azimuth and elevation planes. The
wideband antenna was tested using a carrier wave RF
Figure 2: New "3D" Leica AR25 wideband choke ring
antenna frequency.

The rings of the AR25’s ground plane are arranged The AR25 3D choke ring with its new Dorne-Margolin
with each ring sitting lower than the previous ring so element was compared with an existing high-end 2D
that the choke ring forms a conical shape. The steps choke ring antenna with a standard Dorne-Margolin
between the rings are configured in such a way that element. During the test the Low Noise Amplifier
their length on the longer side is approximately equal (LNA) circuits were bypassed to determine the net
to a quarter wavelength of the lowest frequency gain of the antenna element.
used, in this case L5-L2, and the shorter side is ap-
proximately a quarter wavelength of the highest fre- In order to assess the performance of the antenna
quency used, in this case L1-G1 (Figure 3). The net outside of an anechoic chamber, testing was carried
effect of this configuration is a high impedance sur- out in a ‘real world’ environment by Leica Geosys-
face, which attenuates any surface currents excited tems in Switzerland. The antennas tested were pro-
by the antenna, and eliminates distortion of the am- duction models including the filter and LNA.
plitude and phase pattern. The 3D design improves
low elevation antenna gain while maintaining the Antenna Gain
renowned characteristics of the original choke ring The level of antenna gain is an important indicator of
antenna such as stable phase center, pattern symme- the antenna’s tracking ability. High gain values over
try for amplitude, phase and group delay. the elevation range translate into more complete
data and a higher signal to noise ratio.

While the bandwidth of the AR25 has been signifi-

cantly widened and the low elevation performance

5
optimized, it is clear from Table 2 that the peak an- Table 4: Comparison of the front back ratio for the
AR25 and AT504GG tested at +/-90° elevation
tenna gain at the zenith has not been compromised.

Table 2: Comparison of the antenna peak gain of the AR25 AT504GG

AR25 and AT504GG measured at the zenith (90°El) for
High-Band Frequencies
the three frequency bands 28.9dB 36.4dB
(L1,C1,C2,G1,E1,E2)
AR25 AT504GG
Mid-Band Frequencies
High-Band Frequencies 35.5dB 35.8dB
+4.9dBic +5.5dBic (L2,G2,E6,C6)
(L1,C1,C2,G1,E1,E2)
Low-Band Frequencies
Mid-Band Frequencies 24.9dB 35.6dB
+7.0dBic +7.7dBic (L5,E5a, E5b)
(L2,G2,E6,C6)
Low-Band Frequencies (L5,E5a,
E5b)
+5.3dBic +5.9dBic Antenna Radiation Patterns
The radiation pattern for an ideal antenna would
show consistently high gain from the zenith down to
Table 3 shows that the peak gain on the horizon is
the horizon and would then roll off rapidly for eleva-
better for the AR25 than for the AT504GG across all
tions below the horizon. A consistent radiation pat-
frequency bands, especially the high-band where
tern across all frequencies translates to similar phase
significant improvement is seen. This indicates that
center offset and tracking ability. The greater the
the AR25 has superior low elevation tracking ability.
difference between the Right Hand Circular Polarized
Table 3: Comparison of the antenna peak gain of the (RHCP) and Left Hand Circular Polarized (LHCP) an-
AR25 and AT504GG measured on the horizon (0°El) for tenna gain, the greater the antenna’s resistance to
the three frequency bands reflected signals. The high frequency band antenna
AR25 AT504GG radiation patterns for the AR25 and the AT504GG are
High-Band Frequencies
shown in Figures 4 and 5. The antenna gain values
-4.3dBic -11.1dBic have been normalized to enable direct comparison of
(L1,C1,C2,G1,E1,E2)
the patterns. The peak antenna gain is 0dBic in each
Mid-Band Frequencies
-7.3dBic -9.8dBic case.
(L2,G2,E6,C6)
Low-Band Frequencies (L5,E5a,
-9.0dBic -10.2Bic Figures 4 and 5 show that while the antenna gain of
E5b)
the AT504GG falls away sharply as the high band
signals elevation decreases, the AR25 maintains su-
Front-Back Ratio perior antenna gain over a greater elevation range.
The front-back ratio indicates an antenna’s directivity This translates into superior tracking ability of high
and resistance to multipath (Hekmat et al., 2005). band signals. The peak gain of the AR25 is +5dBic at
The higher the ratio of gain from the front (90° ele- the zenith and -5dBic on the horizon, enabling the
vation) compared to the back (-90° elevation), the antenna to track satellites at all elevation angles. It is
better the antenna’s theoretical ability to reject re- also clear that, while the difference between the
flected signals. The front-back ratio is influenced by a RHCP and LHCP signals is more or less the same for
combination of the antenna’s backside shielding and both antennas for the low elevations, the separation
sensitivity to Left Hand Circular Polarized (LHCP) sig- between the oppositely polarized signals is much
nals. greater for the AR25 for the high elevations, in com-
parison to the AT504GG. This indicates that, not only
Table 4 shows the back-front ratios for the antennas has the AR25 retained the renowned multipath miti-
for each of the 3 frequency bands. While the ratios gation characteristics of the AT504GG at low eleva-
are lower than those for the AT504GG, the values are tions, but has even improved the resistance reflected
still very good and exceed the values from a non- signals at high elevations for the high band frequen-
choke ring antenna. Some trade off is to be expected cies.
due to the significantly improved low elevation track-
ing.

6
 100

Percentage of Expected Observations Received

90

80

70

60

50

40

30

20

10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Elevation [degrees]

AR25 (GPS L1) AR25 (GPS L2)

Figure 4: High-band antenna radiation pattern for AT504GG (GPS L1) AT504GG (GPS L2)
AR25
Figure 6: Completeness of observations by elevation
for the AR25 and the AT504GG

Carrier Phase Multipath Mitiga-

tion
The antenna radiation patterns and front-back ratios
calculated using observations performed in the an-
echoic chamber give us some indication of the an-
tennas ability to mitigate multipath. However, testing
the antenna in a ‘real world’ environment with low
elevation obstructions and nearby reflective surfaces
can show different characteristics.

GPS and GLONASS data was recorded for both the

AT504GG and AR25 antennas every 10 seconds over
Figure 5: High-band antenna radiation pattern for a 24 hour period. The reference antenna was set up
AT504GG
3m away from the subject antenna. This short base-
line length means that, following double difference
Low Elevation Tracking processing, the remaining error will show the residual
The anechoic chamber test results for the AR25 measurement noise. Each epoch of data was proc-
showed significant improvements in peak antenna essed independently resulting in 8640 position solu-
gain on the horizon in comparison to the AT504GG. tions. Figure 7 shows the horizontal coordinate scat-
In theory, this improvement in antenna gain should ter plots for the two antennas while Figure 8 shows
result in superior low elevation satellite tracking. In the height times series. Table 5 shows the standard
order to confirm this theory, ‘real world’ testing was deviations of the both antenna’s.
carried out. Data was recorded down to zero degrees
elevation for both the AT504GG and AR25 antennas
at 1Hz over a 24 hour period.

Figure 6 clearly reveals the AR25’s exceptional recep-

tion of signals from low elevation satellites. For ex-
ample, at 4° elevation the AR25 receives 99.14% of
the expected GPS L1 observations, while the
AT504GG receives just 3.95%. The AR25 displays a
clear superiority in tracking ability from the horizon
up to 10° elevation making it a powerful tool for a
wide range of applications such as such as atmos-
pheric modeling.

7
 elevation of the satellite signal. The difference be-
tween the electrical phase center and the physical
center of the antenna can be removed through Phase
Center Offsets (PCO) and Phase Center Variations
(PCV) calculated through antenna calibration. These
corrections are only effective if the predicted phase
center movement is repeatable for all antennas of
the same model.

The horizontal phase center offsets (HPCO) for the

GPS L1 and L2 frequencies were calculated for 20
production model AR25 antennas in order to asses
the repeatability. Table 6 shows the mean HPCO
values for the L1 and L2 frequencies along with the
respective standard deviations.

Table 6: Horizontal Phase Center Offsets of a sample

of 20 production model AR25 antennas
Figure 7: Horizontal coordinate variation for AR25 (red)
and AT504GG (blue) L1 E L1 N L2E L2N
Mean 0.6 1.0 0.0 0.2
Std. Dev. 0.6 0.4 0.6 0.5

The average HPCO’s are all 1mm or less (Table 6).

However, since the constant offset is removed via
the antenna calibration, the most important factor
for insuring repeatable measurements is the unit to
unit variation. The sub-millimeter standard deviations
show that repeatability of the phase center is very
good.

Conclusion
This paper presents the new AR25 3D wideband
choke ring antenna from Leica Geosystems. With
Figure 8: Height displacement over time for AR25 (red) emerging satellites systems on the horizon, a new
and AT504GG (blue) high performance antenna is needed to encompass
all GNSS signals. The AR25 has sufficient bandwidth
The horizontal scatter for the AR25 is noticeably less to receive all existing and currently planned GNSS
variable (Figure 7), and vertical variations are also signals, while maintaining the highest performance
significantly smaller (Figure 8), than for the AT504GG. standards. A detailed comparison with the renowned
This superiority is confirmed by the smaller standard AT504GG choke ring antenna has shown that the
deviations (STDV) in Table 5. revolutionary new 3D choke ring design, combined
with a new ultra wideband Dorne-Margolin element
Table 5: Dispersion of the calculated coordinates
and high performance LNA, has revealed impressive
STDV (E) STDV (N) STDV (H)
performance improvements, especially with respect
AR25 1.1mm 1.3mm 2.5mm
to low elevation tracking. The reception of the pro-
AT504GG 1.3mm 1.8mm 3.8mm
posed new signals along with additional low elevation
satellites will bring new levels of positional accuracy
Phase Center Variation to reference networks, and benefits the end users of
In an ideal GNSS antenna, the observation point the data. The AR25 has been designed and built for
would correspond exactly with the physical center of durability and will stand the test of time, even in the
the antenna housing. In practice the observation harshest of environments.
point, or electrical phase center, moves around in
three dimensions with the changing azimuth and

8
References
Hekmat, T., Nilass, N. and Maurer, M. (1995), Inte-
grated GPS/GLONASS Antenna for High Performance
Applications, Proceedings of ION GPS-95, September
12-15, Palm Springs, California

IGS (1997), IGS Site Guidelines:

http://igscb.jpl.nasa.gov/network/guidelines/guideline
s.html#allstrict

Kunysz, W. (2001), A Three Dimensional Choke Ring

Ground Plane Antenna, Novatel Inc.
http://www.novatel.com/Documents/Papers/3D_chok
e_ring.pdf

Takac, F. and Walford, J., (2006), Leica System 1200

– High Performance GNSS Technology for RTK Appli-
cations, Proceedings of ION GNSS 2006, September
26-29, Fort Worth, Texas

9
 Whether providing corrections from just a single reference station,
or an extensive range of services from a nationwide RTK network –
innovative reference station solutions from Leica Geosystems offer
tailor-made yet scalable systems, designed for minimum operator
interaction whilst providing maximum user benefit. In full compliance
with international standards, Leica Geosystems’ proven and reliable
solutions are based on the latest technology.

Precision, value, and service from Leica Geosystems.

When it has to be right.

Illustrations, descriptions and technical specifications are not binding and may change.
Copyright Leica Geosystems AG, Heerbrugg, Switzerland, 2009

Leica Geosystems AG
Heerbrugg, Switzerland

www.leica-geosystems.com