The DISAP project was used to develop a GNSS receiver realizing the principle of a synthetic antenna aperture to improve the measurement accuracy. Like for an antenna array, the DISAP receiver combines GNSS signals received at different locations (principle of spatial diversity). The DISAP receiver uses an artificial antenna motion (rotating or vertical) which is measured with sub-millimetre accuracy using magnetic sensors. With proper combination of the signals, the receiver is able to form a synthetic antenna gain pattern, focusing directly towards the satellite and eliminating signals from nearby objects (multipath). This high directivity is very similar (or better) than of receivers using antenna arrays.
Antenna arrays have been studied extensively for decades now and their implementation still poses a series of issues:
- Large number of antenna elements: Apart from being expensive, the effective phase center must be carefully calibrated.
- Multiple RF-paths in the receiver: Each antenna element must be treated separately by the receiver, which implies a separate RF path including down-conversion and analog-to-digital conversion (ADC). Apart from being expensive, these RF paths must be calibrated. Here the ADC is particularly critical and it is very difficult to ensure stability down to carrier phase level accuracy.
The DISAP concept avoids these issues at the expense of requiring an OCXO grade reference clock and moving the antenna. The clock issue is not a serious issue as clocks of the required quality cost about $200 and are the size of a matchbox or smaller. The mechanics of the antenna motion is challenging, but in principle, a solved issue since the invention of radars.
The core idea of the DISAP concept is to generate a synthetic aperture from the displacement of the antenna element. The two key-enablers here are:
- Moving the antenna along a pre-defined path, continuously knowing the exact position of the antenna element.
- A stable reference clock (e.g. an OCXO) allowing coherent combination of the correlation values over a time interval corresponding to one cycle of the antenna movement.
Approximate knowledge of the satellite orbit and accurate knowledge the relative antenna position on its pre-defined trajectory allows the signal processing to combine subsequent correlation values from one satellite coherently, thereby introducing additional phase factors that account for the geometry. These interferometric phase factors cause constructive interference towards the satellite's line-of-sight, while signal contributions caused by multipath are combined in a destructive manner.
Two prototypes have been build. A rotating antenna has a broad range of applications. The vertically moving antenna is optimized to reduce ground multipath.
Extensive test campaigns have shown, that in open sky conditions 50-70 % of the multipath can be removed. In degraded signal environments (e.g. forest), the multipath mitigation is even higher: 60-90%
SETI (prosecution of DISAP)
The primary objective of the SETI project is the development of an innovative receiver technology for improving the performance of future Galileo 2nd generation reference stations for performance monitoring, especially in terms of meeting high accuracy (multipath robustness) requirements even under severe distortion conditions (robustness against spoofing) with one rotating antenna ('synthetic aperture antenna' approach), thus contributing to the Galileo 2nd generation requirements on increased infrastructure (remote Galileo Sensor Stations) robustness against spoofing and on better multipath performance.
Therefore the core objective here is the evaluation of the achievable robustness improvements when using advanced GNSS signal processing algorithms as a trade-off between 'synthetic aperture antenna' compared to the conventional 'fixed antenna' approach. To fully exploit the potential of the new approach, theory development, laboratory tests but also a test measurement campaign are foreseen to evaluate the achievable capabilities under realistic and worst case conditions.
In particular we will:
- Analyze performance requirements w.r.t. to multipath robustness of GNSS reference stations and in particular of G2G sensor stations.
- Provide a vulnerability analysis of G2G sensor stations.
- Propose a synthetic aperture GNSS signal processing scheme capable to detect spoofing signals and mitigate them thereby simultaneously reducing multipath.
- Develop a performance evaluation software for data quality checks of GNSS code/carrier observations. The data evaluation strategy will be motivated by requirements from precise code/carrier based positioning.
- Manufacture and verify two SETI rotating antenna stations (rotating antenna, RF frontend and software receiver) capable of autonomous outdoor operation on a permanent basis.
- Verify the improved multipath performance of SETI stations compared to standard receivers with a fixed antenna. Discuss the impact on improved positioning with the highly advanced method of precise point positioning with ambiguity resolution (PPPAR).
- Design and conduct a spoofing attack using an RF signal generator onto a GNSS receiver with a static antenna within the GATE area. Conduct the same spoofing attack on a SETI station with a rotating antenna. Mitigate the attack and demonstrate that line-of-signals of high quality can be received.
- Discuss the impact of improved multipath and spoofing robustness on GNSS applications in particular on Galileo 2nd Generation Sensor Stations. Discuss the synergy between the synthetic aperture approach and jamming mitigation techniques.
One main benefit of our approach is its ability to identify spoofing signals, even if the receiver already follows them. Receiver internal spoofing detection schemes are mostly only capable to detect the transitions between line-of-sight signals and spoofing signals and are blind once the receiver has been captured. Furthermore, receiver internal anti spoofing algorithms don't provide any mitigation and the receiver has to cease operation.
Another benefit is that we are able to discriminate line-of-sight and spoofing signals by the angle of arrival. Consequently, the spoofing signal can be well mitigated and tracking can continue on the line-of-sight part.
The project was done under ESA contract (2008-2011, ESTEC Ctr. No. 20834)
Consortium: Prime: IFEN, Audens ACT (DE), Uni BW (DE), Telespazio (IT), SEPA (IT)
The positioning capability in severely handicapped environments is the major key issue to use GNSS positioning systems in many areas of application.
Difficult environments are typically (in the order of their current relevance):
- In-car (the success of car/truck GNSS navigation systems today is primarily driven by the capability to have a high-sensitivity GPS receiver, with an antenna inside the car)
- Urban canyon (for car and pedestrian users)
- Indoor (parking garage for cars, buildings for pedestrian, ...)
While the first two environments can be typically covered by today existing High-Sensitivity GPS receivers, the core indoor case is really a challenge towards the physical limits.
The core theme of this project was to push the physical boundaries of GNSS signal acquisition and tracking. The philosophy was to put GNSS at the centre and use other means (IMU, WLAN, A-GNSS, etc) to help increase the GNSS sensitivity. Thus, the emphasis was technology oriented, as opposed to solving a particular problem. This philosophy was also driven by the past experience that navigation systems that depend on terrestrial infra structure (e.g. WLAN, GSM, UMTS etc.) are seldom deployed on a global scale.
In the frame of the DINGPOS project an indoor positioning prototype system making use of GPS/Galileo signals on L1/E1 and L5/E5a in addition to WiFi data, ZigBee data and a MEMS-based IMU dead reckoning algorithm called pedestrian navigation system (PNS) wsas developed. The system uses a partially coherent ultra-tightly coupled integration scheme. The GNSS signal processing (acquisition and tracking) supports coherent integration times up to several seconds, which reduces the squaring loss and mitigates multipath by the high Doppler selectivity of a few hertz.
Extensive tests were performed in the lab using a signal simulator and in the field using GATE (GATE web site). In addition separate test were performed to demonstrate what future technologies (IEEE1588) can contribute in terms of time synchronization. Accurate time is an important aiding instrument for GNSS signal reception.
The results show that it is possible to achieve macroscopic coherent integration times (>1 second), which are the primary enablers if GNSS sensitivity in the order of 0 dBHz or lower is to be achieved.
Six different scenarios (dynamic and static) were defined and tested. The best sensitivity results for acquisition was -3 dBHz, and -8 dBHz for tracking.
The coherent integration while moving the GNSS antenna results in a synthetic antenna aperture. This topic was investigated on a theoretical basis. This principle was further investigated in the frame of a later project, DISAP. The project consortium was lead by IFEN and funded by the EC (FP7).