Session B5: Receiver Design, Signal Processing, and Antennas

The Synthetic Meta-Signal Observations: A Multipath Resistance Assessment
Giovanni Cappello, International PhD Programme / UNESCO Chair “Environment, Resources and Sustainable Development”, Department of Science and Technologies, University of Naples “Parthenope”; Ciro Gioia, European Commission, Joint Research Center; Antonio Angrisano, Department of Engineering, Messina University; Salvatore Gaglione, International PhD Programme / UNESCO Chair “Environment, Resources and Sustainable Development”, Department of Science and Technologies, University of Naples “Parthenope”
Location: Beacon B

Introduction & Literature review
Global Navigation Satellite Systems (GNSSs) are constantly evolving and enhancing in terms of constellations and performance. New services have been introduced to increase GNSS accuracy:Galileo High Accuracy Service, Beidou Precise Point Positioning (PPP) Service Signal PPP-B2, QZSS Multi-GNSS Advanced Orbit and Clock Augmentation - Precise Point Positioning (MADOCA-PPP); and robustness:Galileo Open Signal Navigation Message Authentication (OS-NMA). The new GNSS era foresees also new signals leading to innovative processing solutions. Among all the recent innovations, this study focuses on meta-signal. The meta-signal concept is based on the simultaneous processing of signals on different frequencies, leading to an increased Gabor bandwidth. In the realm of signal processing, researchers in [1] and [2] have put forth the notion of meta-signals. This involves managing two or more signals as a unified element during processing. The integration of these components into a single meta-signal can lead to enhanced tracking capabilities and more precise code measurements. Thanks to the meta-signal approach, the pseudorange precision is increased, leading to a more accurate code-based Position, Velocity, and Time (PVT) solution. There are several ways to process GNSS data using the meta-signal; in that study, the method presented in [2] is used. In particular, the meta-signal observations are synthetically generated at the measurement-level. This approach has been selected because it allows the meta-signal processing with a limited impact on the receiver: in [3], it has been demonstrated that the approach enabled an automotive receiver to process the Galileo Alt-BOC signal with only a firmware update. The synthetic meta-signal approach has been recently developed and it has already been demonstrated that the synthetic meta-signal observations provide enhanced positioning performance [1, 4, 5], especially for Galileo and BeiDou systems [6]. In [7], it has been demonstrated that using a Doble Phase Estimator (DPE) three different measurements are generated for the meta-signal observations: pseudorange, carrier phase and subcarrier phase.
Methodology
The meta-signal carrier-phase measurement is obtained by averaging the carrier-phase of the sideband component, resulting in a Narrow-Lane combination. While the subcarrier phase observation is obtained from the Wide-Lane combination of the side band carrier phases. For the pseudorange two versions can be derived. The first one is a pure code observable, and it is obtained as the average of the pseudoranges of the side band components; this approach slightly reduces the noise with respect to the side band components, but it does not fully exploit the meta-signal. The second one is the high accuracy meta-signal pseudorange obtained by the combination of code and carrier phases from the sideband components. Specifically, the subcarrier is used to smooth the meta-signal pseudorange measurement. The process of the construction of the high-accuracy observable includes the ambiguity resolution of the subcarrier measurements. This is performed by the rounding operation applied on the code-carrier combination closely related to the Hatch-Melbourne-Wübbena combination [8]. The approach provides more accurate pseudorange, but, in the case of incorrect ambiguity resolution, it can cause jumps in the high accuracy pseudorange. The pseudorange jumps are one of the limiting factors for the full exploitation of the synthetic meta-signal approach. In order to fill this gap a pseudorange jump detector, based on the delta-pseudorange, has been implemented and more details can be found in [3]. In this study, an evolution of the detection mechanism has been developed. This allows to have a more robust high-accuracy pseudorange.
The multipath resistance of the synthetically generated measurements is analyzed using the Multipath Linear Combination (MLC), developed by incorporating two significant components: the meta-reconstructed observable as the primary observable and the foremost carrier of the considered GNSS system for reconstructing the Code Minus Carrier (CMC) used in the multipath linear combination. For detecting the cycle slips, a geometry-free combination is used. Furthermore, during the generation of meta-signal observations, a straightforward single-frequency cycle slip detection (singularly for each sideband) is employed based on the selected carrier exploiting carrier-phase and Doppler Shift measurements. This approach provides a robust framework for addressing the challenges posed by multipath effects and cycle slips on meta-signals processing.
Test setup
The available studies on meta-signal have been conducted using specific devices, in open-sky conditions. The application to different type of devices has not been investigated. To fill this gap, several receiver types have been analyzed. This validation has been performed exploiting IGS stations. A total of 32 stations were selected for the analysis and 7 different receiver types were considered. For the assessment, high-rate data from IGS were used, specifically one day of data at 1 Hz were downloaded and processed.
In addition to the effectiveness of the synthetic meta-signal approach using different devices, the analysis focuses on the multipath resistance of the synthetic meta-signal. This assessment is fundamental for navigation in urban scenarios where meta-signals should provide a more accurate solution. The multipath analysis is extended to an urban area; for the urban tests, two different grades of devices are employed: a professional receiver and an automotive grade device. This study allows to fully understand the potentiality of the meta-signal for urban navigation.
The synthetic meta-signal approach is applied to Galileo and BeiDou constellations.
For the Galileo case, the E5 Alt-BOC measurements are generated thanks to the E5a and E5b sideband components. For the BeiDou case, the B2 Ace-BOC measurements are generated thanks to the 3rd generation signals B2a and B2b.
Additionally, for both GNSSs, a simplified version of the synthetical pseudoranges is considered, provided by the mean of the sideband measurements. Multipath distributions and Cumulative Distribution Functions (CDFs) of the multipath error are the evaluated metrics.
Expected results
From the analysis of the above-mentioned metrics, it emerged that the Synthetic Alt-BOC, provides the best multipath resistance when compared with the sidebands. Advantages with respect to the sidebands are found even when the average pseudorange is used. Even for the BeiDou case, it emerged that the benefits of the synthetic measurements with respect to the side-bands components is evident, showing the clear benefit of the reconstructed measurements.
Upon analyzing the 95th percentile of the multipath error distributions for different signal configurations in the Galileo and BeiDou cases, insightful results were obtained. In Galileo case, the E5a signal showed a multipath error of 0.1875 meters, while the E5b signal was slightly higher at 0.1968 meters. The advanced E5 Alt-BOC signal demonstrated appreciable precision with a notably lower multipath error of 0.0956 meters. When the E5 Alt-BOC signal was reconstructed using the average pseudorange, the multipath error increased to 0.1223 meters, but it was still below that of the basic E5a and E5b signals. A further enhancement was observed when the high-accuracy pseudorange was used for reconstructing the E5 Alt-BOC signal, which reduced the multipath error to 0.1071 meters.
Turning to the BeiDou system, the B2a signal exhibited a multipath error of 0.2223 meters, with the B2b signal showing a very similar error of 0.2203 meters. The reconstructed B2 Ace-BOC signal using the average pseudorange yielded a multipath error of 0.1327 meters. This error was significantly improved by utilizing the high-accuracy pseudorange for reconstruction, resulting in a reduced multipath error of 0.1124 meters. These results highlight the efficacy of using high-accuracy pseudoranges in mitigating multipath errors, leading to enhanced signal integrity in both Galileo and BeiDou satellite systems.
As observed, the use of the reconstructed measurements provides better results with respect to the use of the single sideband components, while the original BOC (for the Galileo case) provides the best percentile value.
[1] J.-L. Issler, M. Paonni and B. Eissfeller, "Toward centimetric positioning thanks to L-and S-Band GNSS and to meta-GNSS signals," 2010 5th ESA Workshop on Satellite Navigation Technologies and European Workshop on GNSS Signals and Signal Processing (NAVITEC), pp. 1-8, 2010.
[2] D. Borio and C. Gioia, "Reconstructing GNSS meta-signal observations using sideband measurements," NAVIGATION: Journal of the Institute of Navigation, vol. 70, no. 1, 2023.
[3] D. Di Grazia, F. Pisoni, G. Gogliettino, C. Gioia e D. Borio, «Putting the Synthetic GNSS Meta-signal Paradigm into Practice: Application to Automotive Market Devices,» European Navigation Conference, 2024 (under review), 2024.
[4] P. Das, L. Ortega, J. Vilà-Valls, F. Vincent, E. Chaumette and L. Davain, "Performance limits of GNSS code-based precise positioning: GPS, galileo \& meta-signals," Sensors, vol. 20, no. 8, p. 2196, 2020.
[5] J. Garcìa-Molina, "Unambiguous meta-signal processing: A path to code-based high-accuracy PNT," Inside GNSS, vol. 16, pp. 50-55, 2021.
[6] D. Borio and C. Gioia, "Synthetic Meta-Signal Observations: The Beidou Case," Sensors, vol. 24, no. 1, p. 87, 2023.
[7] M. Paonni, J. T. Curran, M. Bavaro and J. Fortuny-Guasch, “GNSS meta signals: Coherently composite processing of multiple GNSS signals,” Proceedings of the 27th international technical meeting of the satellite division of the institute of navigation (ION GNSS+ 2014), pp. 2592-2601, 2014.
[8] P. J. Teunissen, "Success probability of integer GPS ambiguity rounding and bootstrapping," Journal of geodesy, vol. 72, pp. 606-612, 1998.