Field Performance of A Novel Wideband Optically Transparent GNSS Antenna

Field Performance of A Novel Wideband Optically
Transparent GNSS Antenna
Eray Yasan, Hyok J. Song2, Timothy Talty1 James H. Schaffner2, Duane Carper1, Arthur
General Motors, LLC 1
Bekaryan2
Warren, MI, USA HRL Laboratories, LLC2.
eyasan@msn.com, timothy.talty@gm.com, Malibu, CA, USA
duane.s.carper@gm.com hjsong@hrl.com, jhschaffner@hrl.com,
abekaryan@hrl.com
Abstract—This paper provides and discusses the field test

results on the performance of an optically transparent Global
Navigation Satellite System (GNSS) antenna embedded in the
lamination of the windshield. It has been shown that the location
accuracy by way of transparent antenna is on par with the existing
rooftop production antenna in various environments for the
current applications and use cases.
Keywords—optically transparent antennas; GNSS; GPS,

GLONASS; location; 2D error, reference receiver
I. INTRODUCTION
With the constant advances in infotainment systems in Fig. 1. a) Cross section of windshield lamination b) GNSS antenna design
automobiles, the emphasis on connectivity and the need to have with EM coupled connector
reliable location, there is ever growing increase in the number of
antennas that would support such technologies. Combined with
the challenges in aesthetics and the desire to have minimal
interaction between such antennas have led us to come up with
new design spaces and hence designs. As a result we came up
with a wide-band optically transparent GNSS antenna
embedded in the lamination of the windshield of a vehicle and
demonstrated its effectiveness through field tests against an
existing production antenna. This particular antenna is placed
close to center of the upper part of windshield with the aim of
getting camouflaged by blackout region of the windshield. In
order to assess the location accuracy objectively a commercially
available truth reference system was used. There has been two
different glass manufacturers employed in the fabrication of
such antennas but results are shared from one manufacturer’s
windshield although both have exhibited quite similar
performances.
Fig. 2. Return loss characteristics of transparent GNSS antenna
II. ANTENNA CHARACTERICSTICS
further helped in achieving deeper resonances (< -15dB) in the
Fig 1-a shows the cross section view of the windshield
bands of interest. The vehicle level antenna radiation pattern
lamination where the GNSS antenna is fabricated on PET
measurements showed there is significant field interaction with
substrate using the mesh grid on copper whose details were
the vehicle creating back lobes unlike the rooftop antenna which
described in [1]. The GNSS antenna design is shown in Figure
has no back lobes.
1-b but with copper on a Duroid substrate along with a novel
connector design developed by making use of electromagnetic III. GNSS DRIVE TESTS
coupling. This is a wide band GNSS antenna hence the
connector in the sense that it supports BEIDOU, GPS, A. Test Set-up
GALILEO and GLONASS signals from 1559 MHz to 1607 Fig. 3 shows the test set-up for the road test in order to assess
MHz. Fig. 2 shows the simulated and measured return loss plots the performance of the optically transparent GNSS antenna.
are in good agreement and the presence of the vehicle
978-1-7281-0692-2/19/$31.00 ©2019 IEEE 847 AP-S 2019
Authorized licensed use limited to: VIT University- Chennai Campus. Downloaded on February 23,2022 at 08:36:50 UTC from IEEE Xplore. Restrictions apply.
Novatel© ProPak6 dual frequency (L1, L2) GPS receiver with a current use cases that do not need high accuracy location (e.g.
high grade IMU was used as a truth reference where 2D location sub-meter) requirements.
error was calculated for each case against its location fixes. Two
identical GNSS receivers each capable of simultaneously
acquiring the GPS and GLONASS signals, one connected to
vehicle production sharkfin antenna and the other connected to
transparent GNSS antenna respectively, were used in location
accuracy assessment. A commercially available off-the-shelf
LNA is inserted right after the transparent antenna. While
keeping the cable lengths the same between the antenna
elements and the receivers, the signal strength at the output of
transparent GNSS antenna was adjusted to match sharkfin’s by
use of inline attenuator. Fig. 4. 2D error plots in foliage for each case
Fig. 3. Set-up for location accuracy comparison between the production and the Fig. 5. Average C/No and no. of SV’s used in the solution in foliage
transparent GNSS antenna
B. Location drive tests for performance comparison

TABLE I. 2D error in various environments from drive tests
The road tests were conducted in various signal
environments such as open sky, urban canyon (UC, e.g. Detroit
Downtown), foliage and parking structure in prescribed routes.
In some cases due to the challenge of executing start-stop
operation of instruments, longer routes that have multiple
environments in them were captured and they were separated in
post processing stage accordingly. As an example, Fig 4 shows
2D location error in foliage throughout the duration of the
corresponding route. As can be seen the error characteristics and IV. CONCLUSION
the values are in close agreement with each other for the We have successfully demonstrated through field tests in
production antenna as well as the transparent antenna. Fig. 5 various signal environments that the optically transparent GNSS
provides average Carrier-to-Noise Density ratio (C/No in dB-Hz) antenna embedded in the windshield lamination has a similar
and the number of satellites (SV) used in location solution. location accuracy performance of a rooftop sharkfin antenna.
Although there is almost a constant offset between the C/No‘s of There is work to develop active circuitry for this antenna and to
the two receivers while the one connected to sharkfin being reduce back lobe to help further increase the desired passive
higher, the number of SV’s used in the solution being the same gain.
clearly supports the results of 2D errors not being much different
from each other. The C/No offset could mostly be attributed to ACKNOWLEDGMENT
antenna passive gain and active circuitry gain differences. The authors would like to thank Iqbal Surti from General
Table I tabulates the average, 1-sigma and 2-sigma error Motors and Fred P. Rhine for their role in capturing the field test
values along with the standard deviations for both the sharkfin data and processing the GNSS receiver data.
and transparent GNSS antennas in various environments. It is REFERENCES
observed that the location performance obtained from both
antennas are very similar indicating this new optically [1] H. J. Song, J. H. Schaffner, A. Bekaryan, K. O’Connor, T. Tombs, T. talty,
D. Carper and E. Yasan “Roll-to-roll printed transparent applique
transparent GNSS antenna could be a viable solution for the antennas,” IEEE-2018 APS Conference on Antennas and Propagation,
pp.1671-1672,July2018.
848
Authorized licensed use limited to: VIT University- Chennai Campus. Downloaded on February 23,2022 at 08:36:50 UTC from IEEE Xplore. Restrictions apply.

Field Performance of A Novel Wideband Optically Transparent GNSS Antenna

Uploaded by

Copyright:

Available Formats

Field Performance of A Novel Wideband Optically Transparent GNSS Antenna

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Field Performance of A Novel Wideband Optically Transparent GNSS Antenna

Uploaded by

Copyright:

Available Formats

Field Performance of A Novel Wideband Optically

Transparent GNSS Antenna

Abstract—This paper provides and discusses the field test

Keywords—optically transparent antennas; GNSS; GPS,

978-1-7281-0692-2/19/$31.00 ©2019 IEEE 847 AP-S 2019

B. Location drive tests for performance comparison

You might also like