INTEGRITY ASPECTS FOR DUAL-FREQUENCY DUAL-CONSTELLATION GROUND BASED AUGMENTATION SYSTEM (GBAS)

For navigation in aviation, high performance requirements in terms of integrity, accuracy, continuity, availability, and robustness against interference need to be achieved. The Global Navigation Satellite System (GNSS) alone is not sufficiently accurate and cannot provide precision approach with the required integrity. The Ground Based Augmentation System (GBAS), a development of local-area differential GNSS aims at providing precision approach guidance meeting all of these requirements under low-visibility conditions. GBAS provides differential corrections and integrity monitoring of GNSS, which are broadcast via a Very High Frequency (VHF) radio data link, termed as Very High Frequency Data Broadcast (VDB). The system is currently under development with single-frequency single-constellation Global Positioning System (GPS) L1 stations being already in use. Several studies have shown that ionospheric anomalies that cause large spatial gradients pose a significant threat to this system [1, 2, 3]. With the increasing number of satellites broadcasting signals on a second Aeronautical Radio Navigation Service (ARNS) frequency (L5/E5a) available, the dual-frequency processing becomes a promising option for the next generation aviation users. The use of the signals in a second frequency band can remove the effect of the ionosphere which is one of the dominant sources of errors for the single frequencies user. However, it also yields many different options for processing. Until today there is no clear concept defined on how to use the new signals in a future Dual-Frequency Dual-Constellation (DFDC) GBAS to provide better performance compared to the existing system. This thesis provides essential contributions to the development, standardization and implementation of a future DFDC GBAS. The new service type shall be able to provide sufficient availability, integrity and continuity to be usable globally. Relevant contributors to the error budget are reviewed, characterized and modeled in an appropriate way. The thesis is structured as follows. First, it provides a detailed overview of the existing integrity concept for the single frequency GBAS. The goal of this review is to set the framework and identify the aspects needed in the development of the DFDC GBAS. Next, the thesis discusses the modifications needed when evolving to a DFDC system. The use of the signals from a second frequency band and second constellation enable many different processing modes. In this thesis, the proposal and the selection of the processing modes for a DFDC GBAS is based on upgrading the existing GBAS, but maintaining the main characteristics (e.g. maintaining the existing VDB link). This concept adds several constraints, such as backwards compatibility to the existing single frequency system and the limited remaining VDB capacity. Taking into account all the constraints, the mostpromising processing modes are presented and discussed. Based on the selected modes, the thesis discusses the modifications to the integrity concept and identifies the different aspects that need to be taken into account for each of the processing modes, which is the main contribution on this part. This includes the need for the new ground multipath error models, new airborne multipath models and consideration of the satellite code biases in the integrity concept. Each of these aspects are discussed in the thesis and summarized next including the main contributions. The first error source considered is the multipath and noise errors in the ground reference receivers. Until now, performance analysis of the ground multipath errors using the GPS L1 Coarse/Aquisition (C/A) signal has been studied. In this thesis, the performance analysis of the new signals from the Galileo satellites in the E1 and E5a frequency bands and GPS L5 signals as measured by the experimental GBAS test bed available at Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) is presented. The results show that the raw noise and multipath level of Galileo signals and of the GPS L5 signals are smaller than that of GPS L1. The new signals are also less sensitive to the choice of carrier-smoothing time constant. As the Galileo constellation has different ground track repeatability from GPS, the applicability of the existing methods for the derivation of the ground multipath models is investigated in the thesis. The impact of different effects factor influencing the multipath models such as the smoothing time constant and the receiver parameters are studied. Furthermore, dual-frequency ground multipath error models are derived that differ from the combination of single frequency models. Another important source of error to be considered in a future DFDC GBAS is the airborne multipath. With the removal of the ionospheric error, the airborne multipath error becomes a dominant source of error. The current models are defined only for GPS L1 100 seconds smoothing time constant and no models are available for the new signals broadcast by GPS satellites on L5 band and Galileo satellites. This thesis contributes to the development of the airborne multipath models for the new signals. The work proposes an improved methodology for the derivation of the multipath models for aviation using measurements from flight tests. The improvements include a proposal of a new method for the estimation of the carrier phase ambiguities, the separation of the receiver antenna errors from the multipath errors and an overbounding for the non-Gaussian distribution of the data. The removal of the antenna errors constitutes one of the main changes from the existing models. Until now, the receiver antenna errors were modeled together with the multipath errors. However, the results from this thesis show that these errors have different behavior than the multipath errors and thus a decomposition of these two errors is suggested. Different factor of influence on the airborne multipath, including airframe structure and antenna performance, the receiver bandwidth and correlator spacing and the smoothing time constant are investigated and discussed in this thesis. A third aspect relevant for DFDC GBAS that is discussed in the thesis is the satellite code biases. These biases differ between satellites and depend mainly on the GNSS receiver hardware (e.g. bandwidth and correlator spacing). If the ground reference stations and airborne receiver use different configurations, a residual error remains affecting the GBAS user. These biases can be calculated based on measurements from a high gain antenna. Initial estimates of the satellites code biases are used in this thesis to investigate the impact of the differential satellite code bias on a GBAS user. The results show that the residual errors are not negligible and vary from few centimeters to decimeters. The errors are especially important for the dual-frequency processing modes that combine the errors from different frequencies. Thus, they will affect the GBAS user and need to be taken into account in the integrity budget. The thesis proposes different concepts to include them in the integrity budget of the DFDC GBAS. Finally, the thesis presents an assessment of the expected nominal performance of different processing modes for future DFDC GBAS. The evaluations are based on measurements from flight trials conducted on DLR’s Airbus A320 aircraft. The error models developed in the thesis are used as input for the study. This work contributes to the tradeoff studies needed for a selection of an optimal processing mode for DFDC GBAS