GNSS Indoors: Fighting the Fading, Part 3

Issue Home • Past Issues • Working Papers

Tracking low-power spread spectrum GNSS signals inside buildings is complicated by the effects of various architectures and building materials on signals passing through them. In this final installment of a three-part series, the authors identify and measure some of the key variables affecting the actual Galileo signal-in-space indoors and outline a model for their behavior.

Günter Hein, Andreas Teuber; University FAF Munich

» Download this article (PDF)

Part I
Part II
Part III
In the first part of this series, we extracted some key parameters that characterize the properties of signal propagation under given geometric conditions between transmitter and receiver.

These parameters are arrival rates and amplitudes of individual signals and so-called clusters that are manifestations of the fading phenomena. In order to determine and to quantify these parameters, we deployed a dedicated hardware set-up and applied statistical methods of analysis.

The second part of the series focused on building materials and their dielectric properties. Numerical simulations as well as various test setups enabled us to compute these properties deterministically. Both the analysis of signal transmission and structural factors can be translated into modeling approaches — with a channel model and a transmission model, respectively, as the outcomes.

This third and final part of the series will now describe how the actual Galileo signal-in-space behaves in a particular building to determine whether we can confirm some of the predictions of the aforementioned models.

Galileo Measurements
In this part, we would like to turn to a future scenario: a real Galileo signal that is about to penetrate a building. Readers of the first part of this series may remember the residential-like building located at the town of Berchtesgaden that was used as one of the test buildings for the channel sounder measurements. This structure is also used for our measurements of the Galileo signals.

Although the building is located within the Galileo Test Environment (GATE), which is equipped with Galileo transmitters on the surrounding mountains, for our test a helicopter was deployed in order to cover elevations and azimuths as flexibly as possible.

The receiving equipment consisted of two functionalities: First, the capability to measure the voltage so as to directly infer the attenuation effects of the building materials; second, the capability to record the received signal spectra, which provides insights into how the original signal is distorted. Details on the configuration and the hardware that we employed are given later in the description of the airborne and the indoor test configurations.

Observation Sites
Without doubt, it is meaningful to start the investigations by determining relationships among the various locations of transmitter and receiver. Thus, 15 helicopter positions at six different azimuths and three different elevations were selected.
. . .
Some Initial Conclusions
After analyzing all the parameters under investigation, we can draw the following conclusions regarding the signal power level of a Galileo signal propagating in an indoor environment:
• The signal power level decreases with increasing distance from the furthermost point at which the signal propagation path enters the building.
• As long as there is only one layer of building material to be penetrated (e.g., only a window or a wall), the attenuation gradient only depends on the distance to this uttermost point and is independent of the actual location of the receiver indoors. A first coarse assessment indicates that the signal power decreases by about 3 dB per meter of distance from the exterior wall.
• If two layers of material must be penetrated (e.g., roof and ceiling with an attic room between) the relationship will be more complicated and the gradient will be significantly larger (probably up to 10 dB or more per meter of distance).
• The signal power level decreases with increasing elevation of the signal source. This is inverse to the outdoor situation where tropospheric effects attenuate low-elevation signals.
• With respect to the azimuth, the signal power level is maximal at a direction perpendicular to the closest exterior wall of the building.
• Corners of a building seem to be more favorable entry points than plane walls.
. . .
Conclusion
This three-part column intended to give an overview of the investigation on the properties of GNSS signals propagating indoors. These properties embraced among others the relative geometry of transmitter and receiver as well as the building materials. Finally, after presenting the gathered insights from the developed channel and transmission models in the first two parts, we investigated the received signal power level and the received shapes of the spectra of the true Galileo signal in space.

An analysis based on measurements from 8 to 15 transmitter positions and 3 to 6 receiver positions delivered suitable parameters with which to describe Galileo signal fading satisfactorily. Interestingly, some of the parameters affecting the power level seem to differ from those causing the distortion of the signals.

We attempted to show how suitable information can be obtained from observations collected using dedicated setups. Moreover, it even paved the way to the point of model generation. However, this only provides a suitable description of the phenomena caused by indoor signal propagation.

As for the near future, these models can be useful for simulation purposes. It remains to be determined whether we will ever be able to master the difficult step of implementing the models into a GNSS receiver’s signal-processing chain. We can be sure that a lot more time and effort will be needed before the performance of indoor GNSS will be comparable to its performance outdoors.

Acknowledgment
We would like to thank the German Aerospace Center (DLR) for the funding of the project FKZ 50 NA 0510. Furthermore, we would like to thank Dr. Günter Prokoph and Sebastian Ebenbeck of Work Microwave GmbH for their essential contributions during the Galileo signal field campaign. In particular, we would like to dedicate this work to Sebastian Ebenbeck who tragically died in a recent car accident.

For the complete article, including figures and graphs, please download the PDF at the top of the page.

Issue Home

Manufacturers

The Galileo signal generator developed and constructed by Work Microwave GmbH, Holzkirchen, Germany, was a prototype derived from the one designed in the INDOOR project funded by the German Aerospace Center (DLR). The helical transmitting antenna mounted used on the helicopter — loaned for the tests by IfEN GmbH, Poing, Germany — had been designed within the GATE (Galileo Test and Development Environment) project. The spectrum analyzer used in the indoor environment was an Agilent 8563EC from Agilent Technologies, Santa Clara, California, USA. The deployed receiving antennas were an omnidirectional GPS 704 X pinwheel antenna from NovAtel, Inc., Calgary, Alberta, Canada, which was used to receive all multipath signals at the same time, and an experimental helical beam antenna designed and constructed in the laboratory of Work Microwave GmbH.

Author Profiles

Working Papers explore the technical and scientific themes that underpin GNSS programs and applications. This regular column is coordinated by Prof. Dr.-Ing. Günter Hein. Contact Prof. Hein at <Guenter.Hein@unibw-muenchen.de>.

Andreas Teuber is research associate at the Institute of Geodesy and Navigation at the University FAF Munich. He received his diploma in geomatics engineering from the University of Hannover, Germany. Currently, his main subjects of interest are indoor positioning in general and applying WLAN technology for positioning purposes in particular.