2008:110
M A S T E R ' S T H E S I S
Novel and Robust GBAS Integrity Concepts for Safe Aircraft
Approach Using GPS and Galileo
Pakasit Worracharoen
Luleå University of Technology Master Thesis, Continuation Courses
Space Science and Technology Department of Space Science, Kiruna
2008:110 - ISSN: 1653-0187 - ISRN: LTU-PB-EX--08/110--SE
NOVEL AND ROBUST GBAS INTEGRITY CONCEPTS FOR SAFE AIRCRAFT APPROACH USING
GPS AND GALILEO
DEPARTMENT OF SPACE SCIENCE
LULEÅ UNIVERSITY OF TECHNOLOGY
UNIVERSITÉ PAUL SABATIER TOULOUSE III
A THESIS
SUBMITTED TO THE JOINT EUROPEAN MASTER IN SPACE SCIENCE
AND TECHNOLOGY AND THE COMMITTEE IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE
Pakasit Worracharoen August 2008
ii
Advisor: Dipl.Ing.Boubeker Belabbas, Deutsches Zentrum für Luft- und Raumfahrt Examiner: Anita Enmark, Luleå University of Technology
Examiner: Prof.Dr.Christophe Peymirat, Université Paul Sabatier-Toulousse III
iii
I certify that I have read this thesis and that, in my opinion, it is fully adequate in scope and in quality as a thesis for the degree of Master of Science.
_______________________________________________
Boubeker Belabbas (Principal Adviser)
I certify that I have read this thesis and that, in my opinion, it is fully adequate in scope and in quality as a thesis for the degree of Master of Science.
_______________________________________________
Anita Enmark
I certify that I have read this thesis and that, in my opinion, it is fully adequate in scope and in quality as a thesis for the degree of Master of Science.
_______________________________________________
Christophe Peymirat
Approved for the Joint European Master in Space Science Committee.
iv
ABSTRACT
Ground Based Augmentation Systems (GBAS) using Global Positioning System (GPS) and Galileo is an opportunity navigation system which is used for aircraft precision approach, landing and taxiing with guaranteed accuracy, robustness, integrity, and availability for all weather operations. Galileo supposed to be in operation by year 2013.
The Institute of Communications and Navigation of Deutsches Zentrum für Luft- und Raumfahrt (DLR), as one of the leaders in developing GBAS system, is participating in the project.
The first aim of this thesis is to study the state of the art of Ground Based Augmentation System, Global Positioning System, Galileo, and how they are working by now. We will present the state of the art Global Positioning System (GPS) and give a brief introduction to GBAS, and Galileo constellation. We will give an overview on measurement and error sources. We will study the effect of several kinds of error sources from real data provided by DLR and how these error sources can affect the GPS measurement. We also study the effect of error sources related to the Ground Accuracy Designator (GAD), which plays an important role on GBAS accuracy. The main goal of this study is to select the GAD, in order to be able to broadcast to civil users. Another goal of the work is to implement B- Values and B-Values threshold algorithms for future systems.
The analysis of results will be evaluated and we will conclude the result with respect to the required performances.
Keywords GBAS, Ground Accuracy Designator, B-Values algorithms, Sigma Pseudo- Range ground,
v
ACKNOWLEDGEMENTS
First of all, I would like to thank Sven Molin, SpaceMaster coordinator, for giving me the opportunity to study in Erasmus Mundus SpaceMaster program. Without his support, I could not have got the Erasmus Mundus grant and pursued the SpaceMaster program. He is the person who really made a very big change in my life.
Very special thanks to my supervisor, Boubeker Belabbas, for giving me the opportunity to work on this thesis in Deutsches Zentrum für Luft- und Raumfahrt (DLR). He has helped me in everything. Not only has he helped me choosing work directions, implementing the algorithms, executing the simulation, and analyzing the results but he also helped me finding the resident, driving me back to my apartment, applying for work permit in Germany, and teaching me how to survive in Munich. Working with him has been my great honor.
I am also extremely grateful to my colleagues in communications and navigation department, Anja Grosch, for her support, Patrick Remi for his help and fruitful discussion on the thesis, and all colleagues in department of communications and navigation for friendship and for making me a member of the team. I would like to acknowledge Deutsches Zentrum für Luft- und Raumfahrt (DLR) for its additional financial support on this master thesis work.
I would like to thank all SpaceMaster coordinators, master thesis coordinators, and examiners for their support and understanding, specially: Dr.Victoria Barabash, Dr.Johnny Ejemalm, Prof.Dr.Christophe Peymirat, and Anita Enmark. Without one of them, this thesis could not be accomplished. I also would like to thank all SpaceMaster students for their help and friendship.
I appreciate Dr.Ittibhoom Boonpikum for helping me complete my undergraduate degree and for encouraging me to apply for graduate study.
Finally, I would like to thank my mother – Suweeraya Worracharoen for her endless love, trust, and encouragement. Without her, this thesis work could not have been done.
vi
CONTENTS
NOVEL AND ROBUST GBAS INTEGRITY CONCEPTS FOR SAFE AIRCRAFT APPROACH USING GPS AND GALILEO
I
ABSTRACT IV
ACKNOWLEDGEMENTS V
CONTENTS VI
LIST OF FIGURES IX
LIST OF TABLES XI
GLOSSARY OF ACRONYMS XII
CHAPTER 1
INTRODUCTION 1
1.1THE GLOBAL POSITIONING SYSTEM (GPS) 1
1.1.1ELEMENTS OF GPS 2
1.1.2NAVIGATION SIGNALS 4
1.1.3TRILATERATION 5
1.1.4DIFFERENTIAL GPS 7
1.2GALILEO 8
1.3DLR 9
1.3.1THE DEPARTMENT OF NAVIGATION 9
1.4OUTLINE AND CONTRIBUTIONS 9
CHAPTER 2
THE GROUND BASED AUGMENTATION SYSTEM (GBAS) 11
2.1INTRODUCTION 11
2.2GBASARCHITECTURE OVERVIEW 12
2.3GBASSYSTEM PERFORMANCE 13
2.4THE ACCURACY REQUIREMENTS 13
2.5AVAILABILITY 15
2.6ALERT LIMITS AND TIME TO ALERT 15
2.7CONCLUSION 17
vii CHAPTER 3
MEASUREMENT AND ERROR SOURCES 18
3.1SATELLITE CLOCK AND EPHEMERIS 20
3.2IONOSPHERIC DELAY AND TROPOSPHERIC DELAY 21
3.3MULTIPATH AND RECEIVER NOISE 24
3.4INSTANTANEOUS PSEUDORANGE ERROR (IPRE) 27 3.5OTHER WORKS RELATING TO THE RESULTS FROM YEAR 2003 29
3.6CONCLUSION 30
CHAPTER 4
MULTIPATH AND RECEIVER NOISE ANALYSIS 31
4.1INTRODUCTION 31
4.2THE GBAS GROUND SUB-SYSTEM ACCURACY 31
4.3THE RESULTS 32
4.4CONCLUSION 37
CHAPTER 5
B-VALUES 38
5.1INTRODUCTION 38
5.2B-VALUES 38
5.2.1THE BROADCAST PSEUDO-RANGE CORRECTION (PRC) 39 5.3IMPLEMENTATION OF B-VALUES ALGORITHM 40 5.4SIGMA PSEUDO-RANGE GROUND AND B-VALUES MONITORING 41 5.5IMPLEMENTATION OF B-VALUES THRESHOLD ALGORITHM 42 5.6THE RESULT OF B-VALUES AND B-VALUES THRESHOLD ALGORITHM 43
5.7CONCLUSION 44
CHAPTER 6
6.1SUMMARY OF CONTRIBUTION 45
6.2SUGGESTIONS FOR FUTURE WORK 45
viii APPENDIX A
A.1B-VALUES ALGORITHM 47
A.2 B-VALUES THRESHOLD ALGORITHM 49
BIBLIOGRAPHY 50
ix
LIST OF FIGURES
Figure 1.1: Elements of GPS [Courtesy: The Aerospace Corporation] 2
Figure 1.2: The control segment of GPS 3
Figure 1.3: GPS-Broadcast-Signal [MISENG] 4
Figure 1.4: Trilateration 6
Figure 1.5: Galileo Constellation [Courtesy: ESA] 8
Figure 2.1: GBAS system [ED-114] 12
Figure 2.2: Approach and Landing Operations and Associated Time Intervals [RTCA]
14
Figure 2.3: Protection level and alert limit (source: KN-NA) 16
Figure 3.1: Error Sources in GPS Measurement 18
Figure 3.2: The satellite clock error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
20
Figure 3.3: The Ephemeris error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
21
Figure 3.4: The ionospheric error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
22
Figure 3.5: The tropospheric error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
23
Figure 3.6: The multipath and receiver noise error at L1 frequency for each elevation angle in radians measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
25
Figure 3.7: The multipath and receiver noise error at L1 frequency for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
25
Figure 3.8: The multipath and receiver noise error at L1 frequency for each azimuth angle in radians measured at Oberpfaffenhofen, Germany. Each color
represents one visible satellite.
26
x
Figure 3.9: The instantaneous error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
28
Figure 4.1: The multipath and receiver noise error at L1 frequency for each elevation angle in radians measured at Oberpfaffenhofen, Germany. Each color
represents one visible satellite.
33
Figure 4.2: Multipath and Receiver noise at L1 frequency for each class of elevation angles in degree.
34
Figure 4-3: Relationship between standard deviation of multipath and receiver noise Errors (σpr_gnd) and GAD parameters.
36
Figure 5.1: The flowchart of B-Values algorithm 40
Figure 5.2: The flowchart of B-Values Threshold algorithm 42
xi
LIST OF TABLES
Table 1-1: A summary of errors in GPS measurements [MISENG] 7 Table 2-1: Requirement for Precision Approach and Landing 17 Table 3.1: A summary of the features of ionosphere and troposphere relevant to
GPS signal propagation [MISENG]
24
Table 3.2: Typical pseudorange measurement errors for a single-frequency (L1) receiver [MISENG]
28
Table 4-1: GBAS – GPS Accuracy Requirement Parameters [ED-114] 32 Table 4-2: Relationship between Elevation angle (degrees), amount of sample,
and the standard deviation of multipath and receiver noise errors (σpr_gnd) 34
Table 4-3: GAD parameters with class of Elevation Angle in degrees with 4 installed ground reference receivers
35
Table 4-4: GAD parameters with class of Elevation Angle in degrees with an installed ground reference receivers
35
Table 5-1: σpr_gnd and B-Values monitoring 41
xii
GLOSSARY OF ACRONYMS
DLR German Aerospace Center GAD Ground Accuracy Designator
GBAS Ground Based Augmentation System GNSS Global Navigation Satellite System GPS Global Positioning System
IGS International GNSS Service IPRE Instantaneous Pseudorange error
KN-NA The Institute of Communications and Navigation PRC Pseudo-range Correction
SBAS Space Based Augmentation System UERE User Equivalent Range Error
1
CHAPTER 1
INTRODUCTION
Ground Based Augmentation Systems (GBAS) of Global Positioning System (GPS) is being developed to become the primary navigation system for safe aircraft approach and landing, taxing with high accuracy for all weather operations. Galileo as the newest navigation system developed by European countries and several countries around the world is going to be in operation by year 2013. The institute of Navigation and Communication of the German Aerospace Center as the biggest Aerospace Company in Germany is in charge of developing the GBAS integrity for GBAS system.
The first goal of the thesis is to study the state of the art of GPS, GBAS, and Galileo constellation and understand what we already have. The second goal is to study the effect of several kinds of error sources from real data provided by DLR in year 2003. This knowledge would lead us to choose the Ground Accuracy Designator (GAD) number in order to broadcast to civil users. In addition, chapter 5 of the thesis is to implement the B- Values algorithm and B-Values threshold for DLR, see how they are important for the GBAS system, and how they can be used in the future.
In this chapter, we first present briefly background on Global Positioning System (GPS) including explanation how the position can be determined from GPS signals and differential techniques to become another tool for aiding the aviation navigation. Some information about Galileo, and the institute of Navigation and Communication of DLR would be also explained in this chapter. Finally, contributions, the method of the work, and outline of the thesis are given by the end of the chapter.
1.1THE GLOBAL POSITIONING SYSTEM (GPS)
In the early of 20th Century, when mankind would like to go from one point to another, they can go to the right direction by using a compass, a map, or using the position of the North Star. Even for locating their location in the everyday life was still far beyond what our ancestors could dream of.
Since many years passed, many new technologies have been developed. The Positioning Systems is involved more and more in everyday life. In year 1978, the U.S. Department of Defence (DoD) decided to invest $12 billion [TrimbleGPS] in order to develop a super precise form of worldwide positioning as the very first significant step of Positioning System technology nowadays. The mainly use of GPS at the beginning was to provide
2
accurate information of position, time and velocity for the U.S military forces.
Nevertheless, civilian could also use GPS for their benefits e.g. civil aviation, auto- farming, surveying and mapping, recreation, space navigation etc. Though the information for civilian is not as precise as those for military forces, it is still sufficient enough.
1.1.1ELEMENTS OF GPS
As illustrated in Figure 1.1, GPS is comprised of 3 segments: the space segment, the user segment and the control segment that are working together. The detail of these three segments would be explained as follow.
Figure 1.1: Elements of GPS [Courtesy: The Aerospace Corporation]
- Space Segment
In the beginning, the orbital GPS constellation consists of 24 satellites in 6 circular orbits (increased to 31 satellites in September 2007 for redundancy), that means there are 4 satellites in each orbits with an altitude of 20200 km or 26560 km of the Earth radius at orbital period of 11hr 58min or one-half of a sidereal day. Satellites have an inclination of 55° with respect to the Earth equator. These satellites will provide data messages to user segment.
3 - Control segment
The GPS Control Segment consists of several ground stations located all around the world (see Figure1.2)
Figure 1.2: The control segment of GPS
The master control station (MCS) is located at Colorado Springs, USA. The other ground stations, from the US air force (Hawaii, Cape Canaveral, Ascension, Diego Garcia, and Kwajalein) and also from the National Geospatial-Intelligence Agency (Ecuador, Washington DC, England, Bahrain, Australia, Argentina) as illustrated in Figure 1.2, will transmit received raw data and the received navigation messages to MCS. In addition MCS will analyze, transform, and update all the information which is the important part to user segment in order to estimate position, velocity and time.
- User Segment
Nowadays, user‟s GPS receivers are designed in many models and have been comprised in Electronic devices in many forms e.g. mobile phone, pocket pc, GPS receiver in a car including the newest technology which can fit the GPS receiver in a wristwatch.
In principal, GPS receiver model is a combination of receiver-processor, crystal oscillator and the antenna which can tune the satellite‟s frequencies and receives data messages transmitted by the satellites. Each receiver basically needs 4 satellites in view at minimum to determine the user‟s position and receiver clock bias.
4 1.1.2NAVIGATION SIGNALS
At present, each satellite transmits two radio frequencies band, L1 and L2. The frequencies of L1 centered at 1575.42 MHz and L2 centered at 1227.60 MHz. The precision (encrypted) code (P(Y)-code) is transmitted on both L1 and L2 carrier. The coarse/acquisition pseudorandom noise (PRN) code (C/A code) is transmitted on L1 carrier. As we can see from Figure 1.3, PRN code C(t) would be modulated and also mixed with navigation message on L1 frequency D(t) before it will be broadcasted to civilian. The reason of using PRN code is to make the code which is going to be broadcasted to civilian not as precise as the code for authorized users at L2 frequency.
P(Y) code then can be accessed only by authorized users, C/A code is provided for all civilian.
The GPS C/A-code has a unique sequence length of 1023 bits, called “chips”. The chipping rate of the C/A code is 1.023 MHz made the C/A code to repeat every millisecond [LEE]. Unlike the C/A-code, P(Y)-code is transmitted using Binary Phase Shift Keying (BPSK) modulation at both L1 and L2 frequency and the sequence length of P(Y)-code is much longer (around 104 chips) [LEE] that will make the precision in range measurements much better than that for the C/A-code. The repetition period of P(Y)-code is one week.
Figure 1.3: GPS-Broadcast-Signal [MISENG]
5
Finally, the navigation message (50 Hz) contents of GPS satellite clock corrections, health status, ephemeris parameters, and almanac will modulates the L1-C/A code signal. The bit duration of the navigation message is 20 ms. The GPS receivers will use these transmitted data in order to calculate and determine the position and time.
1.1.3TRILATERATION
In order to understand the measurement and error sources through the thesis, it is essential to understand basic concepts of GPS navigation. This section would briefly explain the principles of wave propagation in the simplest way.
For easy to understand, we firstly consider trilateration that is illustrated in Figure 1.4 as 2- dimensional positioning. Given an observer measures his distance between his position (O) and ground station (S1) using a GPS receiver measures distance using the travel time of radio signals, in this case we exactly knows the position coordinates of S1. The observer position is somewhere on the circle radius r1 centered at S1. Range measurement from the station r2 would also give another circular position. From this point, the observer‟s position could be one of the two points (O and O‟) where the circles intersect.
The observer might be able to eliminate one of these points according to the prior information. Otherwise, the third measurement from the station S3 would be able to determine his position precisely.
6
Figure 1.4: Trilateration
If we put this illustration into a mathematical model, it can be simply determined by solving a set of quadratic equations [MISENG].
(𝑥𝑘 − 𝑥)2+ (𝑦𝑘 − 𝑦)2 = 𝑟𝑘, (1-1)
where
(xk, yk) are the known coordinates of the station S1, S2, and S3 (k = 1, 2, 3), (x, y) are the observer‟s coordinates and
rk is the measured range
This trilateration for GPS navigation simply measures the distance by using the round-trip time of radio signals and very accurate timing. Extension to 3-dimensional positioning is straightforward, in principle. In order to estimate the vertical position, it is required that the elevation angle of at least one of the stations relative to the observer is known. With space-based signal transmitters, each range measurement would determine the position of the surface of a sphere. In other word, intersection of three spheres would be able to determine the point of the observer. The radio signals travel at about 3x108 m/s, and a
S1
S2
S3 r3
r1
r2
O'
‟
O
7
synchronization error of 1µs would result in an error of 300 m in distance measurement [MISENG]. Hyperbolic Positioning and Doppler Positioning will not be explained in this thesis. Readers can get more information from [MISENG].
1.1.4DIFFERENTIAL GPS
Though the basic GPS is the most accurate radio-based navigation system nowadays and already accurate for many operations, there are still some inaccuracies in GPS system that would not be able to correct by normal techniques. The use of differential GPS (DGPS) is toenhance the basic GPS accuracy.
The principal idea of DGPS is to take advantage of the fact that the measurement errors created by a satellite clock, ephemeris, and wave propagation through atmosphere and troposphere are similar for users separated by ten or in the range of hundred kilometers. At the point of the time, these error values would not be much different. In other words, the closer the two users are to each other and the closer the measurement epochs, the more similar the errors would be [MISENG]. Since we can approximate the combined effect of the errors for each satellite and the position of the GPS receiver is known, we can assume that these error estimates can be similar for the GPS users in the area. Each user in the area can apply their measurements to mitigate the errors and improve the quality of the position estimates. This approach is usable for both code and carrier measurements. For navigation, such correction will have to be made available in real time using a radio link. In practice, a user would receive and apply the corrections with some delay, called latency. The closer a user is to the reference station and the shorter the latency, the higher the benefit from the differential corrections [MISENG]. The effect of the error size from basic GPS and DGPS is summarized in Table 1-1 [MISENG].
Source Potential error size (GPS) Error mitigation and residual error (DGPS)
Satellite clock model 2 m (rms) 0.0 m
Satellite Ephemeris
Prediction 2 m (rms) 0.1 m (rms)
Ionospheric Delay 2 – 10 m in zenith direction 0.2 m (rms) Tropospheric Delay
2.3 – 2.5 m in zenith direction at sea level; lower at higher
altitudes
0.2 m (rms) plus altitude effect
Multipath Code: 0.5 – 1 m
Carrier: 0.5 – 1 cm
Uncorrelated between antennas Receiver noise Code: 0.25 – 0.5 m (rms)
Carrier phase: 1 – 2 mm (rms)
Uncorrelated between antennas Table 1-1: A summary of errors in GPS measurements [MISENG]
8 1.2GALILEO
Galileo, a joint initiative of the European Union (EU) and the European Space Agency (ESA), is a GNSS of considerable political, strategic, and economic importance to Europe supposed to be in operation by 2013 [MISENG]. In the Galileo project, not only European countries invested and participated in this project, but China (September, 2003), Israel (July, 2004), India (September, 2005), and South Korea (January, 2006) have also agreements for this investment. In addition, some other countries will join Galileo in the near future. However, the U.S. Government seems not to be interested in this project.
Figure 1.5: Galileo Constellation [Courtesy: ESA]
Although, Galileo is a similar kind of navigation satellite system as GPS and Russian Global Navigation Satellite System (GLONASS), it is subjected to provide more precise measurements than available through those two. The Galileo Constellation is not-to-scale illustrated in Figure 1.5.
9 1.3DLR
DLR is Germany‟s national research center working on Aerospace and Aeronautics. It contains 29 institutes and facilities at 13 locations in Germany. There are also offices in Brussels, Paris, Washington D.C. DLR employs approximately 5700 people and has budget approximately 450 million Euros which is one-third of German government.
1.3.1THE DEPARTMENT OF NAVIGATION
The department of Navigation (KN-NA) is one of the Institutes in DLR working on the Global Navigation Satellite System (Galileo). We mainly work on the integrity in GNSS which focus on ground systems and user terminals. KN-NA also involves with several projects.
- RAIM (Receiver Autonomous Integrity Monitoring) is a technique providing the integrity information from the satellites in view.
- Automatic landing with GBAS is the technique using the simulation in order to demonstrate an automatic landing control system under CAT III conditions using only satellite based navigation.
- Satellite Based Augmentation Systems (SBAS) is also one of the Galileo and GPS integrity systems.
- Multi – Sensor Fusion for High – Dynamic Applications is a development to improve the robustness of GNSS using for automatic landing under all weather conditions using GBAS.
1.4OUTLINE AND CONTRIBUTIONS
Since the contributions of this thesis work will be described later throughout the thesis, this section will briefly summary the outline of this thesis. The second chapter gives a more detailed picture of the GBAS architecture and the GBAS system performance.
Chapter 2 also describes three categories of the accuracy requirements which have been developed until present.
The third chapter explains all kind of concerned error sources, how to reduce the errors and monitor standard deviation of differentially pseudo-range errors in real time at Oberpfaffenhofen, 2003. This chapter also shows the relationship between the standard deviation of multipath and receiver noise at L1 frequency and elevation angle in radians which will be focused in chapter 4.
10
The fourth chapter focuses on the result of the standard deviation of multipath and receiver noise at L1 frequency and elevation angle. This chapter will describe the GAD assessment and its parameters. It explains also how to correct various errors. In the end of chapter 4, it would draw the relationship between standard deviation of multipath and receiver noise errors and GAD parameters. This would lead us for selecting the GAD in order to broadcast to civil users.
The fifth chapter theoretically demonstrates the B-Values, the B-Values technique for GAD assessment, and B-Values monitoring. This will lead us to the implementation of the B-Values and B-Values threshold algorithm in C++. How these algorithms can be used in the future is suggested in this chapter.
Finally, in Chapter 6, the accomplished work is summarized, and the future work directions are suggested in order to improve the obtained results.
11
CHAPTER 2
THE GROUND BASED AUGMENTATION SYSTEM
(GBAS)
2.1INTRODUCTION
GPS is used to determine time, velocity and position. It consisted of one or more satellite constellations and associated receiver and also includes augmentation systems to enable, where necessary, the required performance for the desired operation.
The Ground Based Augmentation System (GBAS) is an augmentation system for navigation system designed to support precision approach operations through the use of terrestrial radio messages. To service those applications, the Federal Aviation Administration (FAA) has been developing the ground based system to augment GPS, namely the Local Area Augmentation System (LAAS). The main purpose of LAAS is to improve the performance of airborne GPS receivers approaching the airport. In Europe, the European Organization for Civil Aviation Equipment (EUROCAE) has been developing high-level performance requirements for a GNSS and GBAS to support precision approach operations. This chapter will present a brief description of the GBAS architecture and introduce the technical terms which will be used to describe the performance of GBAS. It also defines categories of GBAS precision approach and landing based on the level of the system requirements. Finally, we will conclude all the system performance factors involved with GBAS System Performance shortly in a table.
12 2.2GBASARCHITECTURE OVERVIEW
Figure 2.1: GBAS system [ED-114]
As shown in Figure 2.1, GBAS consists of three segments: the Space Subsystem, the GBAS Ground Subsystem, and the GBAS Aircraft Subsystem. The GBAS Signal-in- Space provides the data broadcast from the ground to the aircraft subsystem.
The GBAS ground subsystem includes at least 2 GNSS reference receivers and antennas placed at precisely known locations. With these receivers, the GBAS Ground Subsystem calculates pseudo-ranges for all satellites in view and the ground subsystem broadcasts differential corrections for them based on its own surveyed location. The GBAS aircraft subsystem finally uses this VHF Data Broadcast from the ground subsystem to correct their own pseudo-range measurements for each satellite with the differential correction data received for the GBAS Ground Subsystem. This corrected pseudo-range measurements are used to determine accurately the position of the aircraft relative to the selected Final Approach Path.
The GBAS integrity concept requires the aircraft subsystem to assess the integrity risk due to [ED-144]:
- fault-free, but rare measurement noise,
- faults of one of the ground subsystem reference receivers,
13
taking into account the geometry of the satellites used by the aircraft subsystem. In order to do that, the ground subsystem broadcasts specific integrity data to the aircraft subsystem for each pseudo-range correction. The aircraft subsystem used specific integrity received data to limit the integrity risk.
For the cases where integrity is not a function of current satellite geometry at the aircraft subsystem, such as ranging source failures of ground subsystem faults, the integrity mechanisms are provided by the ground subsystem.
2.3GBASSYSTEM PERFORMANCE
For a fault –free aircraft subsystem, the aircraft subsystem is assumed to have no failure and defined to have nominal accuracy performance. For the GBAS System Performance, we will consider the three criteria as follows [RTCA, ED-144].
- Accuracy: A measure of the difference between the estimated position and the true aircraft position under nominal fault-free conditions. Accuracy is related to the total system (using an onboard receiver with nominal accuracy performances).
- Integrity: A measure of trust that can be placed in the correctness of the information supplied by the total system. Integrity includes the ability of the system to provide timely warnings to the user (alerts) when the system should not be used for the intended operation. Integrity is only related to the Signal-in-Space.
- Continuity: The probability that the system supports Accuracy and Integrity requirements throughout a flight operation without interruption. More specifically, continuity is the probability that the specified system performance will be maintained for the duration of a phase of operation, presuming that the system was available at the beginning of that phase of operation.
These requirements are applied to any GBAS architecture which support the future Category II/III (see Chapter 2.4). In order to meet these requirements, the GBAS will use two or more GNSS signals on different frequencies but that one frequency from one constellation might provide adequate performance with reduced availability.
2.4THE ACCURACY REQUIREMENTS
At present, the accuracy requirements have been developed in three categories as shown in Figure 2.2 and explained as follow [RTCA]. Nowadays, we are in developing of the Category III which is validated against aircraft certification requirements for runway touchdown performance.
14
Figure 2.2: Approach and Landing Operations and Associated Time Intervals [RTCA]
Category I – A precision instrument approach and landing with a decision height not lower than 60 m (200 ft) and with either a visibility not less than 800 m (2400 ft), or a runway visual range not less than 550 m (1800 ft).
Category II – A precision instrument approach and landing with a decision height lower than 60 m (200 ft) but not lower than 30 m (100 ft) and a runway visual range not less than 350 m (1200 ft).
Category IIIa – A precision instrument approach and landing with a decision height lower than 30 m (100 ft) or no decision height and a runway visual range not less than 200 m (700ft).
Category IIIb – A precision instrument approach and landing with a decision height lower than 15 m (50 ft) or no decision height and a runway visual range less than 200 m (700 ft) but not less than 50 m (150ft).
Category IIIc – A precision instrument approach and landing with no decision height and with a runway visual range less than 50 m (150 ft).
15 2.5AVAILABILITY
For GBAS, Availability is the ability of the navigation system to provide the required function and performance which the service can be used for CAT I, CAT II or CAT III in order to provide the reliable navigation information for the aircraft.
The availability is given by a combination of the space subsystem availability and ground and aircraft subsystems availability. The ground and aircraft subsystems loss of availability results from constraints due to Accuracy, Integrity and Continuity of Service requirements. No additional requirement is made on these subsystems related to availability. Availability can be predicted from consideration on statistical performance of the system, considering the effect of system failures. [ED-144]
2.6ALERT LIMITS AND TIME TO ALERT
Alert Limits can be determined in two types which are Lateral Alert Limit and Vertical Alert Limit. In each category, there are alert limits which the error tolerance will not exceed without issuing an alert. The lateral alert limits are given as a function of the horizontal distance of aircraft position from the Landing Threshold Point / Fictitious Threshold Point (LTP/FTP) [RTCA].
The vertical alert limits are given as a function of the height above Landing Threshold Point / Fictitious Threshold Point (LTP/FTP) of aircraft position.
The system Time-to-Alert is the maximum allowable time interval between the beginning of the navigation system being out of tolerance and the appropriate integrity monitoring subsystem providing an alert.
16
Figure 2.3: Protection level and alert limit (source: KN-NA)
Figure 2.3 illustrates the concepts of using the protection level and alert limit in aviation navigation systems. For a safe aircraft, the pilot will be alerted if the position of the aircraft exceeds an alert limit. This alert limit (outer cylinder in Figure 2.3) is defined by the Vertical Alert Limit (VAL) in vertical direction and the Horizontal Alert Limit (HAL) in horizontal direction. If the aircraft locates outside the alert limit without warning, it may be in danger and might hit into an obstacle. The Alert Limit which the pilot needs in real time is defined in table 2-1. Another level illustrated in Figure 2.3 is the protection level.
The protection level indicates how poor the position of the aircraft can be. The Vertical Protection Level (VPL) in the vertical direction and the Horizontal Protection Level (HPL) in the horizontal direction are parameters which are needed to be smaller than the alert limits. The Protection Level will be computed by aviation navigation system and can be varied in size but smaller than the alert limit.
17 2.7CONCLUSION
This chapter provided the background information on GBAS: the GBAS architecture, the system performance (accuracy, integrity, and continuity), and the various types of GBAS precision approaches and landings. We have also briefly learnt about the accuracy requirements developing in 3 categories.
As the last part of this chapter, we will finally conclude all the system performance parameters explained in this chapter in Table 2-1 summarized from standard [ED-144]
and [RTCA].
Operation Accuracy (95% error)
Integrity requirement
Alert Limit (H: Horizontal
V: Vertical)
Maximum Time-to-
Alert
Availability
CAT I
H: 16 m V: 4 m
2 x 10-7 in any one operation
H: 40 m V: 10 m
6.5 s 0.9975
CAT II
H: 6.1 m V: 1.4 m
1 x 10-9 in any one operation
H: 17.9 m V: 4.4 m
2.5 s 0.999
CAT III
H: 3.6 m V: 1.0m to 1.4
m
1 x 10-9 in any one operation
H: 10.4 m V: 2.6 m
2.5 s 0.999
Table 2-1: Requirement for Precision Approach and Landing
From Table 2-1, we can find the current requirements for VAL/HAL for precision approaches. For instance, if the integrity risk or latent failures causes a CAT III user‟s vertical position error to exceed 2.6 meters, the ground facility must detect the event and alert the user within a 2.5 second time-to-alarm. The probability of the ground facility failing in this task should be no greater than 1 x 10-9 per operation.
18
CHAPTER 3
MEASUREMENT AND ERROR SOURCES
In this section, we will consider two kinds of measurement, code phase measurement and carrier phase measurement. The former, known as pseudo-range, is the distance between the satellite and the satellite navigation receiver. Pseudo-range is determined as the transit time so measured multiplied by the speed of light in vacuum [MISENG]. The GPS observation equations for pseudo-range measurements can be determined in the equation (3-1). The latter, carrier phase measurement, is a much more precise measurement than the pseudo-range but an ambiguity in the number of carrier cycles is present. The carrier phase is the difference between the phases of the received phase and the carrier received from a satellite at the instant of the measurement. The equation for code phase measurement relating to the error sources can be defined by the equation (3-2). More detail about code and carrier phase measurements can be found in [MISENG].
Figure 3.1: Error Sources in GPS Measurement
19 The measurements are determined from [MISENG]
𝜌 = 𝑅𝑚𝑛 + 𝑐 𝑏𝑚− 𝐵𝑛 + 𝐼𝑚𝑛 + 𝑇𝑚𝑛 + 𝑀𝑚𝑛 + 𝜈𝑚𝑛, (3-1) 𝜙 = 𝑅𝑚𝑛 + 𝑐 𝑏𝑚 − 𝐵𝑛 + 𝐼𝑚𝑛 + 𝑇𝑚𝑛 + 𝑁𝑚𝑛𝜆 + 𝑝𝑚𝑛 + 𝜀𝑚𝑛, (3-2) where
ρ is the measured code phase measurement, or pseudo-range, ϕ is the measured carrier phase measurement,
𝑅𝑚𝑛 is true range from satellite n to receiver m, 𝑏𝑚 is the receiver clock bias (offset from GPS time), 𝐵𝑛 is the satellite clock bias (offset from GPS time), I is the ionospheric delay,
T is the tropospheric delay,
M, p are multipath errors respectively in code and phase measurements, N is the integer ambiguity,
𝜆 is the carrier wavelength (for L1 frequency, 𝜆𝐿1 = 𝑓𝑐
𝐿1 ≈19 cm), 𝜈 represents other code phase measurement errors and
𝜀 represents other carrier phase measurement errors.
As we can see in (3-1) and (3-2), GPS measurements are subject to various factors of errors. In this thesis, we will group the error sources in three groups regarding to the sources of error which will be shortly explained in the following sections.
The GPS broadcast signal (see Figure 1.3) will be transmitted from the satellite through the atmosphere. The delay will be generated though this region. After signals have propagated through the atmosphere, some of the signals would also bounce to the buildings, forest, and other bouncing surface. This would generate multipath errors. Other errors would be created by the satellites, instruments, and the environment around them.
All errors will be discussed in this chapter.
20 3.1SATELLITE CLOCK AND EPHEMERIS
The ephemeris and clock data are constantly broadcasted by the satellites and estimated by the control segment. Receivers, which maintain an almanac of this data for all satellites, compare, compute, and update hourly its almanac to the estimated data coming in.
Figure 3.2: The satellite clock error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
The satellite clock which is the difference between the actual clock and the satellite clock generated by the similar procedure, introduces about 1-2 m error in the root mean square sense. However, there are some clock errors for which the magnitude is higher than 5 meters as illustrated in Figure 3.2.
These errors can be reduced by increasing the data uploaded frequency to the satellites.
21
Figure 3.3: The Ephemeris error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
According to the difference between the actual position and velocity of the satellite, the satellite ephemeris error is generated associated with both the estimation of the current values of the data and the data predicted by the broadcast ephemeris model. This error is normally around 1-2 m of magnitude in the root mean square (rms) sense as illustrated in Figure 3.3.
3.2IONOSPHERIC DELAY AND TROPOSPHERIC DELAY
The second type of error is the error of wave propagation through dispersive media. This is the source of ionospheric delay and tropospheric delay. The ionosphere is a region of ionized gas formed along the layer of the atmosphere ranging in altitude from 50 to 500 km. These ionized particles affect the speed of GPS signals propagation from a satellite to a receiver. The characteristics of the ionosphere also change between day and night. The error will also change with the eleven-year solar cycle.
22
Figure 3.4: The ionospheric error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
As we can see from Figure 3.4, the ionospheric error is depending on the season change and the solar cycle. The value of the error is mostly in the positive side. Changes in ionospheric delay can be measured accurately with the carrier phase measurements while the carrier is tracked continuously. The carrier and its modulating signal (i.e., the code and navigation message) propagate at different speeds through the ionosphere: the code phase is delayed while the carrier phase is advanced by the same amount [MISENG]. However, we cannot compare the data for the whole eleven year solar-cycle since there is no more information and data available at this time.
Another concern in this kind of error is tropospheric delay. The troposphere is the lower part of the earth‟s atmosphere extending from the earth‟s surface to 9 km above the poles and 16 km above the equator. It mainly consists of water vapor and dry gases (N2 and O2) and varies in temperature and pressure. These dry gases and water vapor refract GPS signals and present an additional delay. From Figure 3.5, we found that the tropospheric error is completely different from the ionospheric error. The reason is that the troposphere, unlike ionosphere, is non-dispersive for GPS frequencies (i.e., the refractive index does not depend upon the frequency of the signal). The speed of propagation of GPS signals in
23
the troposphere is lower than that in free space and, therefore, the apparent range to a satellite appears longer depending on the satellite elevation angle. The phase and group velocities are the same, and the measurements of code and carrier at L1 and L2 frequencies all experience a common delay. This delay cannot be estimated from GPS measurement. However, the troposphere does not cause a big error. It can be corrected by using atmospheric models [MISENG].
Figure 3.5: The tropospheric error for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
As we can see from Figure 3.5, the tropospheric error also follows seasonal dependencies.
We found that the error is larger between September and November of the year.
One possibility to reduce errors from ionospheric delay and tropospheric delay is to compare the relative speeds of two different signals. This technique is called "dual frequency" measurement [TrimbleGPS] which is normally used in advanced receivers.
Finally, we summarize the main feature of GPS signal propagation through the ionosphere and the troposphere in Table 3-1.
24
Ionosphere Troposphere
Variability High (diurnal, seasonal, and solar cycles; solar flares)
Low (significant change only in the wet component, which is less than 10% of the delay
Zenith delay Several meters to several tens of meters
2.3 – 2.6 m at sea level
Obliquity factor (= 1 in zenith direction)
≈ 1.8 at 30º elevation angle;
2.5 at 15º; 3 at 5º
≈ 2 at 30º elevation angle; 4 at 15º; 10 at 5º
Modeling error for zenith delay
1 – 10 m, or more 5 – 10 cm (without meteorological data Dispersive for GPS
frequencies?
Yes (therefore, the delay can be measured from dual- frequency measurements)
No
Table 3-1: A summary of the features of ionosphere and troposphere relevant to GPS signal propagation [MISENG]
The multiplier of the zenith delay is called obliquity factor. The propagation path length of a signal through the ionosphere increases with the zenith angle. The increased path length is accounted for in terms of obliquity factor [MISENG].
3.3MULTIPATH AND RECEIVER NOISE
Multipath error occurs when the transmitted signal from the satellite bounce off various surfaces before it travel to the navigation receiver. Multipath error normally affects code measurements around 1-5 m error and 1-5 cm rms error for carrier measurements [LEE].
However, the advanced receiver can reduce this kind of error by considering only the earliest signals coming in, which are the direct one. The remaining error is receiver noise.
The receiver noise affects also the code and carrier measurements. This receiver noise is generated in many parts of receiver: the antenna, amplifiers, cable, etc. The receiver noise will affect the final signal especially when the signal-to-noise ratio is low. It introduces around 0.5 m of pseudo-range error and 1-2 mm of carrier phase measurement error.
25
Figure 3.6: The multipath and receiver noise error at L1 frequency for each elevation angle in radians measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
Figure 3.6 is raw data of multipath and noise errors as function of elevation angle at L1 frequency. We notice that the more elevation angle it is, the less error we will get.
Figure 3.7: The multipath and receiver noise error at L1 frequency for the year 2003 measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite From Figure3.7, we notice that there are more multipath and receiver noise errors in the end of the year. The reason would be that the weather is worse and there might be snow, rain, and water vapour that will affect the receiver.
26
Figure 3.8: The multipath and receiver noise error at L1 frequency for each azimuth angle in radians measured at Oberpfaffenhofen, Germany. Each color represents one visible satellite.
From Figure 3.8, the different of error for each azimuth angle are not easily distinguished, however we notice that the error generated by azimuth angle is normally around 1 m and might be very high to 10 m of magnitude.
Multipath and receiver noise error will be discussed more in Chapter 4.
