Department of Science and Technology Institutionen för teknik och naturvetenskap
Linköpings Universitet Linköpings Universitet
([DPHQVDUEHWH
LITH-ITN-KTS-EX--02/26--SE
7HFKQLFDO9HULILFDWLRQDQG
9DOLGDWLRQRI$'6%9'/0RGH
IRU(QURXWH$LUVSDFHDQG0DMRU
7HUPLQDO$UHDV
Petter Granberg & Roger Li
2002-12-18
LITH-ITN-KTS-EX--02/26--SE
7HFKQLFDO9HULILFDWLRQDQG
9DOLGDWLRQRI$'6%9'/0RGH
IRU(QURXWH$LUVSDFHDQG0DMRU
7HUPLQDO$UHDV
([DPHQVDUEHWHXWI|UWL.RPPXQLNDWLRQVRFK
WUDQVSRUWV\VWHPYLG/LQN|SLQJV7HNQLVND+|JVNROD
&DPSXV1RUUN|SLQJ
Petter Granberg & Roger Li
Handledare: Harald Malén och Tobias Andersson
Examinator: Peter Värbrand
5DSSRUWW\S Report category Licentiatavhandling P Examensarbete C-uppsats P D-uppsats Övrig rapport _ ________________ 6SUnN Language Svenska/Swedish P Engelska/English _ ________________ 7LWHO Title
Technical Verification and Validation of ADS-B/VDL Mode 4 for En-route Airspace and Major Terminal Areas
)|UIDWWDUH $XWKRU
Petter Granberg & Roger Li
6DPPDQIDWWQLQJ
Abstract
This report is a technical verification and validation of Automatic Dependent Surveillance – Broadcast (ADS-B) over Very High Frequency Data Link Mode 4 (VDL Mode 4) for the use as surveillance in terminal areas and en-route airspace in non-radar areas. The main objective is to verify that ADS-B/VDL Mode 4 fulfils the technical requirements for an implementation at Kiruna airport, Sweden. Comparison has been made to the current requirements for Secondary Surveillance Radar (SSR).
The work in this report has been conducted in three phases: preliminary study, tests and verification and validation. During the preliminary study documents primarily from EUROCONTROL and ICAO were used to find out which requirements that were applicable. The next part consisted of both practical tests and theoretical verification of the VDL Mode 4 performance. Finally the results from the tests were validated and put together in this report.
Main conclusion from this report is that ADS-B/VDL Mode 4 fulfils the corresponding SSR requirements. Therefore ADS-B/VDL Mode 4 should be able to serve as primary mean for surveillance in non-radar areas.
The results from this report will constitute a part of the technical subset of future safety case for ADS-B in non-radar areas. The complete safety case will be used to authorize ADS-B/VDL Mode 4 in non-radar airspace, both in Sweden and internationally.
,6%1
BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB ,651/,7+,71.76(;6(
_________________________________________________________________
6HULHWLWHORFKVHULHQXPPHU,661
Title of series, numbering ___________________________________
1\FNHORUG
Keyword
ADS-B, Accuracy, ATM, Availability, CAA, Continuity, Coverage, En-route Airspace, GNSS, GPS, Integrity, Kiruna, Latency, Luftfartsverket,
2002-12-18
85/I|UHOHNWURQLVNYHUVLRQ
Division, Department
Institutionen för teknik och naturvetenskap Department of Science and Technology
Acknowledgements
We would like to thank all the people at the Swedish Civil Aviation Administration (SCAA) and at Linköping University, which in some way has been involved in the work with this report, for their support and assistance. Special thanks to Christian Axelsson and Harald Malén, supervisors at SCAA; to Tobias Andersson and Peter Värbrand, supervisor and examiner respectively, at Linköping
University. We would also like to thank Matts Eriksson and Jonas Lundmark, also writing their master thesis at the SCAA, for the exchange of thoughts and ideas.
Finally, thanks to Niclas Gustavsson at the SCAA for giving us the opportunity to perform this master thesis at the SCAA.
Petter Granberg & Roger Li Norrköping, December 2002
Programme/Area: TEN-T/DG TREN Project Number: 2001/EU/SE/GR-5003
Project Title: North European ADS-B Network Update Programme, Phase II (NUP II)
Document Id: SCAA_NUP_WP33_TVV_ADS-B_ER_TMA_1.0 Internal Reference: N/A
Version: Version 1.0
Work package: Work Package 33
Date: 2003-01-13 Status: Released Classification: Public
Author(s): Petter Granberg and Roger Li/SCAA (petter.granberg@lfv.se, roger.li@lfv.se).
Technical Verification and Validation of
ADS-B/VDL Mode 4 for En-route Airspace
and Major Terminal Areas
Document Identification
Programme: TEN-T/DG TREN Project Number 2001/EU/SE/GR-5003
Project Title: North European ADS-B Network Update Programme, Phase II Project Acronym NUP II
Chairman of Steering Committee Mr. Bo Redeborn, SCAA +46 1119 2388
bo.redeborn@lfv.se
Project Technical Manager Mr. Niclas Gustavsson, SCAA +46 1119 2273
niclas.gustavsson@lfv.se
Partners Swedish Civil Aviation Administration, SCAA NAVIAIR
Finnish Civil Aviation Administration, FCAA TERN
Norwegian Air Traffic and Airport Management, NATAM Scandinavian Airline Systems, SAS
Lufthansa, Deutsche Lufthansa, DLH Deutsche Flugsicherung GmbH, DFS DGAC/STNA
Airbus France Austro Control ADS-B Scatsta
EUROCONTROL Experimental Centre, EEC Belgocontrol
AVTECH Sweden AB
Document title Technical Verification and Validation of ADS-B/VDL Mode 4 for En-route Airspace and Major Terminal Areas
Document Id SCAA_NUP_WP33_TVV_ADS-B_ER_TMA_1.0 Organisation Internal Reference N/A
Work Package No Work Package 33 Version Version 1.0
Status Released Classification Public
Date 2003-01-13
Author(s) Petter Granberg and Roger Li/SCAA (petter.granberg@lfv.se, roger.li@lfv.se). Organisation maintaining document SCAA
File SCAA_NUP_WP33_TVV_ADS-B_ER_TMA_1.0 Printed 2003-01-13
Abstract
This is a technical verification and validation of ADS-B/VDL Mode 4 for the use in Kiruna Airspace, Sweden. The purpose of this document is to support ATC implementations of ADS-B, both in Sweden and internationally.
Distribution List
Name Organisation
ANDERSSON, Tobias Linköping University
AXELSSON, Christian SCAA
DANIELSON, Lars SCAA
ERIKSSON, Matts SCAA
GRANBERG, Petter SCAA
GUSTAVSSON, Niclas SCAA
HASSLAR, Göran SCAA
JAKOBSSON, Sven Swedish Aviation Safety Authority
KÅRHED, Örjan Linköping University
LI, Roger SCAA
LUNDMARK, Jonas SCAA
MALÈN, Harald SCAA
ODIN, Owe SCAA
VÄRBRAND, Peter Linköping University
Control Page
This version supersedes all previous versions of this document.
Version Date Author(s) Pages Reason
0.1 2002-11-20 Petter Granberg, Roger Li/SCAA
All First draft
0.2 2002-12-04 Petter Granberg,
Roger Li/SCAA All Rewrite after review 0.3 2002-12-11 Petter Granberg,
Roger Li/SCAA Several Rewrite after review 0.4 2002-12-13 Petter Granberg, Roger Li/SCAA Section 6, Appendix C Minor changes 1.0 2003-01-13 Petter Granberg,
Table of Contents
Executive Summary ...12 1. Introduction...13 1.1. DOCUMENT OBJECTIVES...13 1.2. AUDIENCE...13 1.3. REVISIONS...13 1.4. SCOPE OF DOCUMENT...13 1.5. METHOD...132. Definitions and explanations...14
3. ADS-B/VDL Mode 4 Concept ...17
3.1. CONCEPT OF ADS-B ...17
3.2. CONCEPT OF GROUND-BASED AUGMENTATION SYSTEM (GRAS)...17
3.3. ADS-B RELATIONSHIP TO PRIMARY AND SECONDARY SURVEILLANCE RADAR...17
3.4. DESCRIPTION OF VDL MODE 4...18
3.5. GROUND NETWORK...20
4. Kiruna airspace environment description ...22
5. VDL Mode 4 performance ...24 6. Accuracy...25 6.1. ACCURACY REQUIREMENTS...25 6.2. VERIFICATION OF ACCURACY...26 6.3. VALIDATION OF ACCURACY...28 7. Availability...33 7.1. AVAILABILITY REQUIREMENTS...33 7.2. VERIFICATION OF AVAILABILITY...34 7.3. VALIDATION OF AVAILABILITY...35 8. Integrity...36 8.1. INTEGRITY REQUIREMENTS...36 8.2. VERIFICATION OF INTEGRITY...36 8.3. VALIDATION OF INTEGRITY...36 9. Continuity ...37 9.1. CONTINUITY REQUIREMENTS...37 9.2. VERIFICATION OF CONTINUITY...37 9.3. VALIDATION OF CONTINUITY...37 10. Latency ...38 10.1. LATENCY REQUIREMENTS...38 10.2. VERIFICATION OF LATENCY...39 10.3. VALIDATION OF LATENCY...40 11. Update rate ...41
11.1. UPDATE RATE REQUIREMENTS...41
11.2. VERIFICATION OF UPDATE RATE...41
11.3. VALIDATION OF UPDATE RATE...41
12. Coverage ...42
12.1. COVERAGE REQUIREMENTS...42
12.2. VERIFICATION OF COVERAGE...42
12.3. VALIDATION OF COVERAGE...43
13. Failure Mode, Effects and Criticality Analysis (FMECA) ...44
13.1. SCOPE OF FMECA...44
13.2. CONTENTS OF THE FMECA ...44
13.3. FMECA FOR ADS-B/VDL MODE...46
14. Monitoring ...51
15. Conclusions ...54
15.1. RECOMMENDATIONS FOR FURTHER WORK...54
15.2. POSSIBLE SOURCES OF ERROR...55
Appendix A - Test procedures ...56
Appendix B - Detailed SSR requirements ...72
List of Figures
Figure 1-1. Scope of document within dotted circle. ...13
Figure 3-1. GRAS service scenario ...17
Figure 3-2. Schematic picture and photo of a VDL Mode 4 transceiver...19
Figure 3-3. Schematic picture and photo of ground station rack (front view)...20
Figure 3-4. Ground network logical design...21
Figure 4-1. Kiruna TMA ...22
Figure 6-1. Flight path for accuracy test inside Kiruna TMA. ...27
Figure 6-2. Flight path for accuracy test in en-route airspace in the Kiruna area. ...28
Figure 6-3. Plotted ADS-B and RTK position reports inside Kiruna terminal area...29
Figure 6-4. ADS-B and RTK position reports in the part of the flight where the tracks deviates the most. ...30
Figure 6-5. ADS-B, SSR and RTK position in a part of the flight where the three tracks clearly deviates. ...31
Figure 6-6. SSR and ADS-B barometric altitude vs RTK geometric altitude. ...32
Figure 7-1. Static transceiver test, Arlanda. ...34
Figure 10-1. Explanation of the terms data age and total latency...40
Figure 12-1. SSR coverage requirements...42
Figure 13-1. Scope of the FMECA ...44
Figure 15-1. Data transfer and processing...55
Figure 15-2. Flight test, Kiruna TMA...57
Figure 15-3. Flight test, Kiruna en-route airspace ...59
Figure 15-4. Flight test, Arlanda TMA...64
Figure 15-5. Flight test, Arlanda en-route airspace ...66
Figure 15-6. Static transceiver test at Arlanda airport. ...69
Figure 15-7. Explanation of azimuth bias and slant range bias. ...72
Figure 15-8. Systematic horizontal errors due to azimuth bias 0.1°...73
List of Tables
Table 6-1. Horizontal positional accuracy requirements ...25
Table 6-2. ADS-B position error inside terminal area...29
Table 6-3. ADS-B position error in en-route airspace. ...30
Table 6-4. Geometric altitude error inside terminal area...32
Table 6-5. Geometric altitude error in en-route airspace. ...32
Table 7-1. Radar Surveillance Data Availability. ...33
Table 7-2. ADS-B data availability requirements...34
Table 7-3. Location of the transceivers used in static test at Arlanda airport. ...35
Table 7-4. Availability in flight test in Kiruna and at Arlanda. ...35
Table 7-5. Availability in stationary transceiver test at Arlanda airport...35
Table 9-1. Continuity in flight test in Kiruna and at Arlanda. ...37
Table 9-2. Continuity in stationary transceiver test at Arlanda airport...37
Table 10-1. ADS-B latency requirements for different phases of flight ([Ref 19] appendix K, table K-1). ...38
Table 10-2. The data age value in ADS-B reports translated to milliseconds...39
Table 13-1. FMECA elements ...45
Table 13-2. FMECA for ADS-B/VDL Mode 4 ...50
Table 14-1. Monitoring methods...53
Abbreviations
ACARS Aircraft Communications Addressing and Reporting System ADS Automatic Dependent Surveillance
ADS-B Automatic Dependent Surveillance – Broadcast
ATC Air Traffic Control
ATM Air Traffic Management
ATS Air Traffic Services
CAA Civil Aviation Administration
CAPS Centralised Access Point Server
CNS Communication, Navigation and Surveillance CPDLC Controller Pilot Data Link Communication
CRC Cyclic Redundancy Check
DCAA Danish Civil Aviation Administration (SLV)
DFS Deutsche Flugsicherung GmbH
DGNSS Differential Global Navigation Satellite System DGPS Differential Global Positioning System
DLH Deutsche Lufthansa
DME Distance Measuring Equipment
EU European Union
EUROCAE European Organisation for Civil Aviation Electronics FEC Forward Error Correction
FIS-B Flight Information Services - Broadcast
FL Flight level
FMECA Failure Mode, Effects and Criticality Analysis FTP File Transfer Protocol
GNSS Global Navigation Satellite System
GPS Global Positioning System
GRAS Ground-based Regional Augmentation System
GSC Global Signalling Channel
ICAO International Civil Aviation Organisation IEC International Electrotechnical Commission
ILS Instrument Landing System
LAPS Local Access Point Server LFV Luftfartsverket (SCAA)
LSC Local Signalling Channel
MAEVA Master ATM European Validation Plan
MASPS Minimum Aviation System Performance Standards
MDT Mean Down Time
MSL Mean Sea Level
MTBCF Mean Time Between Critical Failures
N/A Not Available
NATAM Norwegian Air Traffic and Airport Management
NEAN North European ADS-B Network
NEAP North European CNS/ATM Application Project NM Nautical Miles (1NM = 1852 metres)
NUC Navigation Uncertainty Category
NUCp Navigation Uncertainty Category for Position
NUP NEAN Update Programme
PSR Primary Surveillance Radar RAPS Regional Access Point Server
RTCA (inc.) Requirements and Technical Concepts for Aviation
RWY Runway
SAS Scandinavian Airline Systems
SCAA Swedish Civil Aviation Administration (LFV) SLV Statens Luftfartsvæsen (DCAA)
SSR Secondary Surveillance Radar
STDMA Self-organising Time Division Multiple Access
SWR Standing Wave Ratio
TDMA Time Division Multiple Access TEN-T Trans-European Network – Transport TIS-B Traffic Information Service – Broadcast
TMA Terminal Area
UDP User Datagram Protocol
UTC Universal Time Co-ordinated
VDL VHF Digital Link
VHF Very High Frequency
VIP VDL Mode 4 Interface Protocol WGS 84 World Geodetic System 1984
References
[Ref 1] “ADS-B in VDL Mode 4” (version 0.5), SCAA_NUP_WP33_ADS-B in VDL Mode 4, Christian Axelsson/Swedish CAA, 2002
[Ref 2] “Analysis techniques for system reliability – Procedure for failure mode and effects analysis (FMEA)”, IEC 812, CEI (Commisson Electrotechnique Internationale), 1985
[Ref 3] “Analytical and Experimental Observation of Ionospheric and Tropospheric Decorrelation Effects for Differential Satellite Navigation during Precision Approach”, Christie et al/Stanford University, 1998
[Ref 4] “Automatic Dependent Surveillance Requirements”, ADS/SPE/CR-TF-REQ/D1-08, EUROCONTROL, September 28, 2001
[Ref 5] “Driftsäkerhet och underhåll (andra upplagan)”, ISBN: 91-44-39111-0, Karl-Edward Johansson /Studentlitteratur, November 1996
[Ref 6] “Environmental Conditions and Test Procedures for Airborne Equipment”, ED-14D, EUROCAE, 2000
[Ref 7] “EPIC 8 Cyclic Redundancy Checks and Data Integrity”, Roland Rawlings/EUROCONTROL
[Ref 8] “Failure Modes, Effects and Criticality Analysis (FMECA) for VDL Mode 4” (Draft), CNS-DD-484, Tommy Bergström/CNS System, 2001
[Ref 9] “Functional specification: NUP Ground Network”, 65046-0100-08, Bengt-Arne Skoog/Aerotech Telub, April 22, 2002
[Ref 10] “Geodetic calcuations”, http://www.auslig.gov.au/geodesy/datums/calcs.htm, Geoscience Australia, access date August 12, 2002
[Ref 11] “GRAS Service Description”, SCAA_NUP_WP34_GRAS Service Description_1.0, Ottmar
Raeymaeckers, Gunnar Frisk and Abdul Tahir/Swedish CAA, August 19, 2002
[Ref 12] “Ground-based Regional Augmentation System (GRAS) based on VDL Mode 4”, Ottmar
Raeymaeckers, Gunnar Frisk and Abdul Tahir/Swedish CAA, 2000
[Ref 13] “Interim Minimum Operational Performance Specification for VDL Mode 4 Aircraft Transceiver for
ADS-B”, ED-108, EUROCAE, 2001
[Ref 14] “Line of Sight Calculator”, http://www.rebelcbsales.com/calcs/distcalc.html, Rebel CB Sales, access date August 13, 2002
[Ref 15] “Manual of Air Traffic Services Data Link Applications (First Edition)”, Doc 9694-AN/955, ICAO,
1999
[Ref 16] “Manual of the Secondary Surveillance Radar (SSR) Systems”, Doc 9684-AN/951, ICAO, 1997
[Ref 17] “Manual on Implementation of a 300 m (1000 ft) Vertical Separation Minimum Between FL 290
and FL 410 Inclusive”, Doc 9574-AN/934, ICAO, 1992
[Ref 18] “Math Forum”, http://mathforum.org/library/drmath/view/51879.html, Drexel University, access date July 15, 2002
[Ref 19] “Minimum Aviation System Performance Standards for Automatic Dependent Surveillance
[Ref 20] “Minimum Operational Performance Specification for Secondary Surveillance Radar Mode S
Transponders”, ED-73A, Eurocae, February 1999
[Ref 21] “Operational Services and Environment Definition (OSED) ADS-B Kiruna Application”,
SCAA_NUP_WP24_0.6, Owe Odin/SCAA, March 1, 2002
[Ref 22] “Preliminary Safety Assessment report”, NUP-1-2K-FR-PSA-SN-001, Airsys Navigation Systems
GmbH, October 22, 2001
[Ref 23] “Radar Surveillance in En-Route Airspace and Major Terminal Areas”,
SUR.ET1.ST01.1000-STD-01-01, EUROCONTROL, March 1997
[Ref 24] “Sannolikhetsteori och statistikteori med tillämpningar”, ISBN 91-44-03594-2, Gunnar
Blom/Studentlitteratur, 1989
[Ref 25] “Validation framework for the NUP Phase II” (Draft 0.4) EEC_NUP_WP25_02-0.4, Blandine
Lemaire, Eric Hoffman, 2002
Executive Summary
This report is a technical verification and validation of Automatic Dependent Surveillance – Broadcast (ADS-B) over Very High Frequency Data Link Mode 4 (VDL Mode 4) for the use as surveillance in terminal areas and en-route airspace in non-radar areas. The main objective is to verify that ADS-B/VDL Mode 4 fulfils the technical requirements for an implementation at Kiruna airport, Sweden. Comparison has been made to the current requirements for Secondary Surveillance Radar (SSR).
The work in this report has been conducted in three phases: preliminary study, tests and verification and
validation. During the preliminary study documents primarily from EUROCONTROL and ICAO were used to
find out which requirements that were applicable. The next part consisted of both practical tests and theoretical verification of the VDL Mode 4 performance. Finally the results from the tests were validated and put together in this report.
Main conclusion from this report is that ADS-B/VDL Mode 4 fulfils the corresponding SSR requirements. Therefore ADS-B/VDL Mode 4 should be able to serve as primary mean for surveillance in non-radar areas. The results from this report will constitute a part of the technical subset of future safety case for ADS-B in non-radar areas. The complete safety case will be used to authorize ADS-B/VDL Mode 4 in non-radar airspace, both in Sweden and internationally.
1. Introduction
1.1. Document objectives
The aim of this report is to show that ADS-B/VDL Mode 4 at least fulfils the current requirements for Secondary Surveillance Radar (SSR) to be used by air traffic controllers as provider of aircraft surveillance information. In some cases official (ICAO or EUROCONTROL) requirements already exist for ADS-B performance, in these cases the report will verify if these requirements are fulfilled by ADS-B/VDL Mode 4. In other cases, where official requirements only exist for SSR, the report will verify whether ADS-B/VDL Mode 4 fulfils the current SSR requirements or not. The findings of the report will constitute one of the technical parts of a future safety case that will be used by the Swedish Aviation Safety Authority to authorize the use of ADS-B/VDL Mode 4 for air traffic surveillance.
1.2. Audience
The audience for this report is the partners in the NUP phase II and the Swedish Aviation Safety Authority.
1.3. Revisions
This document is maintained by the Swedish Civil Aviation Administration – SCAA.
1.4. Scope of document
The issues that are addressed in this report are the performance of ADS-B/VDL Mode 4 when used from aircraft to ground station, and vice versa, in en-route airspace and major terminal areas. Most of the findings will also be valid for the aircraft to aircraft segment. Performance is expressed in terms of availability, accuracy, integrity, continuity and latency. Issues related to other parts of the surveillance system, for example data transfer to Air Traffic Control (ATC) or display of surveillance data at ATC have not been addressed in this report.
ICAO Annex 10 volume III chapter 6 VHF Air Ground Digital Link (VDL) contains requirements about
frequency usage, power transmitted from ground and aircraft installations as well as physical layer and link layer protocols and services. These issues are not covered by this report but can still be of interest.
Figure 1-1. Scope of document within dotted circle.
1.5. Method
This study will be executed in three phases, Preliminary study, Tests and Verification and Validation. Documentation of all the activities will be made throughout the duration of the study. The validation framework established for validation activities within NUP II, [Ref 25], based on the MAEVA project work, has been used throughout the project.
ATC
GROUND STATION AIRCRAFT
2. Definitions and explanations
Accuracy Accuracy is a statistical measure of performance that describes how well a measured value agrees with a reference value [Ref 26].
Aerodrome A defined area on land or water (including any buildings, installations, and equipment) intended to be used either wholly or in part for arrival, departure and surface movement of aircraft.
Availability Availability is the probability that a system can provide the required function and performance within the specified service volume at the start of an intended operation [Ref 26].
Continuity Continuity is the probability that a system will continue to provide the required function and performance during the intended period of operation, assuming it was available at the start of the operation [Ref 26].
En-route Airspace En-route airspace is the volume of airspace outside terminal areas (see below), where the climb, cruise and descent phases of flight take place and within which various types of air traffic services are provided ([Ref 23] Section 3.1.2).
EPOS Epos is a DGPS service that is available with ±10 metre accuracy – claimed with 95% confidence. There are twelve Epos service GPS Reference Stations, co-located with Swedish Land Survey sites throughout Sweden, which generate local differential corrections.
EUROCONTROL EUROCONTROL is the European Organisation for the Safety of Air Navigation. It was founded in 1960 for overseeing air traffic control in the upper airspace of the member states. The most important goal today for EUROCONTROL is the development of a coherent and co-ordinated air traffic control system in Europe. Currently there are 29 member states.
Flight Level (FL) Flight level is the name for the pressure altitude reported by an aircraft. It is measured in hundreds of feet, FL100 = 10 000 feet
Global Signalling Channel
(GSC) The Global Signalling Channels are the two worldwide channels/frequencies (same frequencies all over the world) used for transmitting and receiving messages via VDL Mode 4. In complement to the GSCs, Local Signalling Channels can be used (see below).
Integrity Integrity is the probability that errors will be detected. For example a correct message must not be indicated as containing one or more errors, or a message containing one or more errors may not be indicated as being correct [Ref 15].
Latency Latency can be described as the elapsed time between a system input and the corresponding system output ([Ref 4] Section 6.6).
Local Signalling Channel
(LSC) A Local Signalling Channel is to serve as a complement to the Global Signalling Channel (see above) but with only local coverage, primarily close to major airports.
Major Terminal Area A major terminal area is the volume of airspace surrounding one or more principal aerodromes (see above). The lateral extent will vary, depending on the disposition of aerodromes within, or adjacent to the terminal area. The vertical dimensions will vary with the way the airspace and the procedures for handling the air traffic flow is organised ([Ref 23] Section 3.1.2).
Mode A, C and S The identification of an aircraft is reported by its SSR transponder (see below). This is called the Mode A code and it is manually set by the pilot on request of Air Traffic Control (ATC). Special codes are reserved for emergency situations (7500 = hijack, 7600 = radio failure, 7700 = general emergency).
With Mode C the altitude of an aircraft, expressed in flight levels (FL, see above), is reported by its SSR transponder. Possibly values are within FL10 and FL1267
Mode S is an extension of the wide-spread SSR protocol to include the possibility of sending longer messages between ATC and pilots.
Primary Surveillance Radar (PSR)
Primary Surveillance Radar is a system where a radar ground station transmits interrogation signals and calculates a distance to surrounding aircraft hit by the signal. This calculation is based on the time it takes for the echoing signal to return to the radar ground station. No equipment on-board the aircraft is needed.
Real Time Kinematic (RTK)
GPS Real Time Kinematic GPS is a high precision GPS equipment. At a well-defined reference position a station uses the carrier wave transmitted from the GPS satellite to calculate a very precise distance to the satellite. This calculation is then used by the aircraft GPS receiver to serve as correction to the position calculated from the time received from the GPS satellites.
Secondary Surveillance Radar
(SSR) Secondary Surveillance Radar is a system where a radar ground station transmits interrogation signals to aircraft transceivers. The aircraft transceiver replies and the time it takes for the reply to reach the ground station is used to calculate the distance. The aircraft transceiver reply also contains other information, e.g. ID and/or altitude, depending on which modes the SSR supports.
Terminal Area (TMA) See “Major Terminal Area”, the term major is used when the terminal area surrounds one or more major airports. Around medium or smaller airports the term Terminal Area is used.
3. ADS-B/VDL Mode 4 Concept
This section is intended for readers who are not familiar with the concept of ADS-B/VDL Mode 4. Other readers can skip this section. The contents of this section are excerpts from the VDL Mode 4 Master Document [Ref 26], Ground-based Regional Augmentation System (GRAS) in VDL Mode 4 [Ref 12], GRAS Service Description [Ref 11], and Minimum Aviation System Performance Standards for Automatic Dependent Surveillance Broadcast (ADS-B) [Ref 19]. The text has been somewhat shortened and edited.
3.1. Concept of ADS-B
Automatic Dependent Surveillance Broadcast (ADS-B) is a surveillance application that via a broadcast mode data link transmits parameters such as four-dimensional position (x, y, z and time), track and ground speed, aircraft or vehicle identification and additional data as appropriate. ADS-B is automatic because it transmits automatically without any need for external stimulus; it is dependent because it relies on on-board navigation sources and on-board broadcast transmission systems to provide surveillance information to other users. Any user, either aircraft or ground-based, within the range of this broadcast, may receive and process ADS-B surveillance information. The data to be transmitted is derived from on-board navigation and position-fixing systems, often, but not necessarily, Global Navigation Satellite Systems (GNSS) such as GPS and GLONASS. The transmissions take place at specified intervals for utilisation by any air and/or ground users requiring it ([Ref 26] Section 3.5.5 and Annex B).
3.2. Concept of Ground-based Augmentation System (GRAS)
GRAS provides a GNSS augmentation service, primarily to enhance position accuracy, over a wide geographical area (or “region”) through a network of ground stations. GRAS is made up of multiple ground stations; placed at well-defined positions, with overlapping coverage, typically located at airports, see Figure 3-1. Since the position of the ground station is known the station serves as a reference point. Each station broadcasts locally valid augmentation messages through a data link.
Figure 3-1. GRAS service scenario
3.3. ADS-B Relationship to Primary and Secondary Surveillance Radar
Primary Surveillance Radar (PSR) is generally described as independent, non-cooperative surveillance. PSR is independent, because the user of the surveillance information derives the surveillance data on the subject aircraft by his/her own means (i.e., illumination of the target). PSR is non-cooperative, because the subject aircraft is not required to carry any commonly standardized surveillance-related equipment. Although the nature of the derived information is somewhat limited for ATS support (e.g. poor blip-scan-ratios, confusion with clutter, and lack of target altitude and ID), PSR is robust in the sense that surveillance outage failure modes are limited to those associated with the ground radar system.Except for barometric altitude, Secondary Surveillance Radar (SSR) is considered to be an independent, cooperative surveillance system. With SSR, the interrogating ground-based radar station makes the target range and bearing estimates based on the interrogation and subsequent reply of a cooperating transceiver
Ground network
GRAS Broadcast
Remote service monitoring, etc
.
[Ref 11]carried by the aircraft. Mode-S augments SSR with an aircraft addressing and two-way data link capability. SSR provides more detailed information than PSR, but SSR does not provide information on unequipped aircraft or where the aircraft component has failed. For this reason, SSR considerations must include availability of the aircraft transceiver as well as the interrogating ground station.
ADS-B is a dependent cooperative surveillance system. As with SSR, common equipage is required for participation in the system. In addition to position information, ADS-B provides other types of information, including velocity and aircraft intent. Unlike ground-based radar, ADS-B information is directly available to airborne as well as ground receivers. ADS-B surveillance failure modes, however, include both the navigation source of the reported state vector information as well as operation of the ADS-B transmitter units or the ADS-B receiver units.
3.4. Description of VDL Mode 4
VDL Mode 4 is a VHF data link that can be used for time-critical applications, providing digital communications between mobile stations (aircraft and airport surface vehicles) and between mobile stations and fixed ground stations. It was developed for CNS/ATM aviation applications, including broadcast applications (e.g. ADS-B) and point-to-point communications (e.g. Controller Pilot Data Link Communication, CPDLC). The most prominent property of VDL Mode 4 is its efficient exchange of short repetitive messages. VDL Mode 4 transmits digital data in a standard 25 kHz VHF communications channel and employs a Time Division Multiple Access (TDMA). A TDMA system divides the communication channel in time segments by first specifying a frame, the term superframe is used for VDL Mode 4, which in turn is subdivided into time
slots. The duration of a VDL Mode 4 superframe is one minute and each superframe contains 4500 time
slots (i.e. 75 time slots per second). The start of each slot is an opportunity for a station to transmit.
A surveillance system such as ADS-B cannot have restrictions that specify the maximum number of participating stations. Therefore VDL Mode 4 is capable of handling overload situations (i.e. more slots are required than currently available) and to adapt to the traffic in a controlled and safe manner. This adaptation is accomplished by the VDL Mode 4 transceiver being able to detect overload situations. When there are too few time slots available, the transceiver decreases its operating range; slots reserved by distant users can then be “re-used”.
Built on the Self-organising TDMA (STDMA) concept, the unique feature of VDL Mode 4 is the way that the available transmission time is divided into a large number of short time-slots synchronised to UTC. Each time slot may be used by a radio transceiver (mounted on aircraft, ground vehicles or at fixed ground stations) for transmission of data. The exact timing of the slots and planned use of them for transmissions are known to all users in range of each other, in that way the data link can be used efficiently and users do not transmit simultaneously. As a result of this ‘self-organising’ protocol, VDL Mode 4 does not require any ground infrastructure to operate and can therefore support air-air as well as ground-air communications and applications.
3.4.1. VDL Mode 4 transceiver and ground station
Figure 3-2. Schematic picture and photo of a VDL Mode 4 transceiver.
The hardware components in VDL Mode 4 can be divided into three major groups; mobile transceiver, ground station (including ground station transceiver) and ground network. The mobile VDL Mode 4 transceiver contains a GNSS receiver, a communication processor and a VHF transceiver, see
Figure 3-2.
The major difference between the mobile transceiver and the ground station transceiver is that the ground station transceiver (named VRS in Figure 3-3) does not contain a GNSS receiver, instead the ground station transceiver gets the GNSS information from an external GNSS receiver placed in the ground station, see Figure 3-3. Other vital parts of the ground station, except for the already mentioned transceiver and GNSS system, are:
• Time Reference Subsystem (TRS), provides the VDL Mode 4 transceiver and DMS (see below) with
precise time information.
• Data Management Subsystem 1 (DMS1), the “intelligent” part of the ground station processing GRAS
data.
• Data Management Subsystem 2 (DMS2), same as above but processing ADS-B data and other
services such as TIS-B and FIS-B.
In addition the ground station also contains units for power supply and power back up (UPS) in case of loss of power.
GNSS Antenna VHF
Figure 3-3. Schematic picture and photo of ground station rack (front view).
3.5. Ground network
The ground network consists of three logical levels, see Figure 3-4. At the lowest level the ground system can be found, consisting of one or more ground stations connected to a Local Access Point Server (LAPS), typically located at an airport. Ground systems in the same region will be connected to the same Regional Access Point Servers (RAPS) and regions close to each other will be connected to the same Centralised Access Point Server (CAPS). At present there are two LAPS in Sweden, one at Kiruna airport and one at Arlanda airport. With only two LAPS running there is no need to accommodate the three logical layers in three different physical layers; accordingly the software of RAPS currently runs in the CAPS and thus the ground network only contains two physical layers. As the number of ground stations increases, the RAPS
and the CAPS will be separated into different physical layers. The CAPS is currently located at Swedish CAA HQ in Norrköping but the future location will probably be at Arlanda.
Figure 3-4. Ground network logical design
The ADS-B data flow from LAPS to CAPS can be described as follows ([Ref 9] Section 2.6.1):
1. Data is received in the LAPS from one or more ground stations. Duplicate data is filtered. Data is then passed to the RAPS
2. Data is received in the RAPS from one or more LAPS and duplicate data is filtered. Data is then passed to the next level CAPS
3. Data is received in the CAPS from one or more RAPS and duplicate data is filtered. Data is then passed to other CAPS that have requested it.
Authorized parties, such as air traffic control and airlines, get access to ADS-B data by placing a subscription at a CAPS. The CAPS get their ADS-B data from one or more LAPS or RAPS. For instance, this means that ADS-B data received in the Kiruna ground station is forwarded to the LAPS located at Kiruna airport. The ADS-B data will then be sent from the LAPS to a CAPS located at Arlanda (via a RAPS). The Kiruna tower (TWR) has a subscription for certain data, the CAPS processes the data according to this subscription and then sends it back to be presented to the air traffic controllers in the Kiruna TWR.
CAPS
Ground Station
NUP Ground Network for ADS-B data
CAPS RAPS RAPS LAPS LAPS Ground Station RAPS LAPS LAPS Ground system
4. Kiruna airspace environment description
Most tests and trials within this study will be conducted in the Kiruna area in northern Sweden, where a VDL Mode 4 ground station is up and running since late spring 2002. The main part of the following description of Kiruna airspace is taken from the NUP phase II document Operational Services and Environment Definition
ADS-B Kiruna Application [Ref 21].
The situation today is that aircraft operate in non-radar covered airspace and must report to an air traffic controller every 15 minutes by voice, the controller must also acknowledge the report by voice. The use of ADS-B will enable radar-like services that will result in more efficient traffic handling. The air traffic controller will have superior situational awareness of the traffic in the Kiruna area compared to the present (no radar or display at all). From a pilot perspective the workload in the cockpit will decrease in the future since there is no need to report position and altitude as this data is autonomously transmitted via the VDL Mode 4 data link.
Kiruna terminal area (TMA) extends from 3100 ft MSL up to FL95 and from 4500 ft MSL up to FL95 in the southern part. The horizontal extension can be seen in Figure 4-1 below.
Figure 4-1. Kiruna TMA
Sundsvall Control has radar coverage down to FL80 in the southern part of Kiruna TMA and down to FL130 over the airport. Below FL80 radar coverage is not available.
VHF coverage is good with a little reservation for the area northeast of the airport. With the mountainous terrain in the Kiruna area, communication via VHF can sometimes be difficult for aircraft at low altitude due to the line-of-sight limitation that characterises VHF radio.
The traffic characteristics are a mix of scheduled, demanded and general aviation traffic. Occasionally military traffic is crossing the control zone and the terminal area. The main traffic flow is southbound. Except for the radar shortage, there are currently no capacity problems and none are expected in the future.
5. VDL Mode 4 performance
As stated in Section 1.1, Document objectives, the purpose of this report is to verify that ADS-B via VDL Mode 4 can be used for surveillance in terminal area and en-route airspace. VDL Mode 4 performance will be measured in terms of accuracy, availability, continuity, integrity and latency, Section 7 to 10 deals with these performance measures. Issues concerning update rate and coverage can to some extension be found in Section 11 and 12.
The first step in the validation work has been to find out which are the requirements that need to be fulfilled in order to use ADS-B/VDL Mode 4 for surveillance. The primary source of information has been documents from ICAO and EUROCONTROL (for a complete list of references see page 10). In some cases requirements for ADS-B cannot be found and in these cases requirements for SSR have been used, where applicable.
In order to verify the VDL Mode 4 performance three major tests have been performed, one in the Kiruna area and two at Arlanda airport, Stockholm.
The main purpose of the Kiruna test was to perform flights in the terminal area and en-route airspace to investigate the altitude accuracy and the position accuracy in the ADS-B reports. This test and the result of the test are more thoroughly described in Section 6.2, Verification of accuracy, and Section 6.3, Validation of accuracy. After having analysed the log files from the Kiruna flights it was found that no differential corrections (GRAS) had been received. Further analysis showed that a bug in the transceiver software caused the lack of GRAS. After the software manufacturer had released a new version of the software, a new test had to be performed. This second accuracy test was performed in the Stockholm area with the Arlanda ground station providing differential corrections.
The third test was a long time test with two stationary transceivers placed at Arlanda airport. The main purpose of this test was to provide large amounts of log data, primarily to be able to examine availability, continuity and latency. More information about this test can be found in Section 7.2.
During all tests log data have been recorded after being processed in the ground station but before being processed by the local access point server (LAPS).
6. Accuracy
6.1. Accuracy requirements
6.1.1. Horizontal accuracy requirements
A current operational requirement for SSR data is that the horizontal positional accuracy shall have an error distribution with a root mean square (RMS) value equal to or less than 500 metres for en-route airspace and equal to or less than 300 metres for major terminal areas ([Ref 23] Section 5.2.3). This should be interpreted as follows: from the true position, a standard deviation error of 500 metres and 300 metres respectively is allowed, see Appendix C.4. ADS-B shall also be able to fulfil these requirements.
Major Terminal Area En-route Airspace
Error (RMS value) 300 metres 500 metres
Table 6-1. Horizontal positional accuracy requirements
Note: Requirements above only concerns horizontal position, altitude is not taken into account.
In the same document [Ref 23] as referred to above the technical requirements for an individual SSR sensor can also be found ([Ref 23] Section 6.3.3.1). The technical requirements for SSR accuracy are however difficult to transfer into requirements for ADS-B due to large differences between the two techniques. Since the SSR horizontal accuracy requirements are not transferable to ADS-B, the complete SSR requirements are accounted for in Appendix B - Detailed SSR requirements. Some knowledge about the magnitude of the allowed horizontal errors in a SSR system is given by the brief description of the technical requirements below.
The horizontal accuracy of a SSR can be described in terms of azimuth (or bearing) accuracy and range accuracy. Maximum allowed azimuth bias for a SSR is 0.1 degrees, i.e. the direction to the target reported by the SSR may not differ from the correct direction with more than 0.1 degrees. An azimuth bias results in an error in metres that increases with the distance between the target and the radar station.
In addition to the azimuth bias requirement there are also requirements concerning range errors. The range is the distance between the target and the radar station. Somewhat simplified, the maximum allowed range error for a SSR can also be seen as a function of the distance between the target and the radar station, the longer distance – the larger error. For example, a range error of one metre per nautical mile (between the target and the radar station) is allowed for a SSR.
When validating the ADS-B/VDL Mode 4 accuracy performance, the operational requirements will be used as the requirements that VDL Mode 4 shall be able to fulfil.
6.1.2. Altitude accuracy requirements
A SSR transceiver gets the aircraft altitude from a pressure altimeter onboard the aircraft and this is also the case with an ADS-B/VDL mode 4 transceiver. Accordingly there is no difference in altitude accuracy between SSR and ADS-B/VDL mode 4 when using barometric altitude. However ADS-B/VDL mode 4 also includes the possibility to use the geometric altitude that has a higher accuracy than the barometric altitude. The test described in Section 6.2.1, Test method for accuracy of position and altitude, will show the magnitude of the difference is between the barometric altitude and geometric altitude.
6.1.3. Time stamp accuracy requirements
In the ground station a time stamp is attached to each position report. For SSR data the time stamp error shall be less than 100 ms ([Ref 23] Section 6.3.3.1), or in other words, the ground station clock shall have an
Definition
Accuracy is a statistical measure of performance. Accuracy describes how well a measured value agrees with a reference value [Ref 26].
accuracy of 100 ms or better. ADS-B shall be able to fulfil the same requirement according to ([Ref 4] Section 6.2).
6.2. Verification of accuracy
6.2.1. Test method for accuracy of position and altitude
In surveying it is very difficult to determine an exact “true” value. A true value can however be closely approximated using very accurate measuring equipment and techniques. These close approximations can then be used for estimating the accuracy of other survey systems. To measure the accuracy of position and altitude for VDL Mode 4, tests have been performed in the Kiruna area and the Arlanda/Stockholm area using high-precision RTK (Real Time Kinematic) equipment as a reference. The original intended use for RTK is in geodetic positioning. RTK is used by the SCAA’s flight inspection primarily when calibrating
Instrument Landing System (ILS) equipment at airports. ILS is an approach and landing aid that provides
continuous, reliable and highly accurate position information to the aircraft. Today ILS is the standard system to enable precision approaches.
GPS receivers calculate its position by measuring distances from the GPS satellites. The most common method, used by all receivers, is to calculate the difference between the time a signal is transmitted from a satellite and the time it is recorded by the receiver, using the code embedded in the satellite's signal. This measurement is called code phase. GRAS used in VDL Mode 4 uses code phase to calculate the distances to the satellites. The true position of the ground station is known, and therefore a correction can be calculated for each satellite in view. These corrections are broadcasted to all aircraft within range of the ground station. RTK is based on GPS and also has a ground based reference station sending out differential corrections to other users. To calculate the corrections to the position provided by the GPS satellites the RTK reference station uses both code phase and carrier phase to calculate the distances to the satellites. Carrier phase serves as a compliment to code phase measurement by measuring the carrier phase of the satellite carrier wave. This method provides centimetre-resolution.
The RTK equipment has four different levels of accuracy for the reported position: 0 (no corrections), 1 (EPOS), 2 (float) and 3 (fixed) where level 3 has the best accuracy, ±1 centimetre for a stationary target. The RTK accuracy level depends on the distance from the reference station. For a moving target the accuracy also varies as a function of the velocity. During the tests the aircraft ground speed was about 200-250 kts and the RTK report rate was 10 reports per second, this gives a horizontal error of about 10-12 metres. The vertical error remains small though, around 1 centimetre, since the altitude changes are very small over a short period of time. The “stationary” error and the “velocity” error should thus be added to get the real accuracy. Accuracy level 2 and 1 have an accuracy of about 10 metres for a stationary target. The error due to the target moving remains the same as in level 3, i.e. 10 metres at 200 kts. Accuracy level 0 means that no corrections are received, only GPS position without any corrections.
More information about the flight test and the equipment used can be found in appendix A.1.2
The flight tests described in this section were performed in the Kiruna area September 23 2002 and in the Arlanda/Stockholm area November 9 2002. The tests follow appendix A.1.1 Test plan, which was created to facilitate for pilots and observers. The idea was to perform one flight in Kiruna terminal area (TMA) and one flight in en-route airspace in the Kiruna area. When analysing the result from the Kiruna test a software bug was discovered causing the differential corrections (GRAS) not to function as intended. After this bug had been fixed the Arlanda test was conducted to verify that GRAS worked as intended. During these flights log data were collected from the RTK transceiver, the aircraft VDL Mode 4 transceiver and at the output from the ground station to the ground network. This log data has been compared and differences in position and altitude have been calculated. During the comparisons the data recorded by the RTK have been used as the “true” value.
6.2.1.1. Accuracy test inside terminal area
The tests inside Kiruna TMA and Stockholm TMA were performed close enough to the reference station for the RTK reports to have the best accuracy, level 3. During the flight both ADS-B data via VDL Mode 4 and RTK data were recorded. Then the two logs were compared to find out how much the ADS-B position differed from the true position, i.e. the position recorded in the RTK log.
Kiruna TMA extends from 3100 ft up to FL 95 (9500 ft). The recording of RTK data started when the aircraft was close to the airport (see Figure 6-1) and above 3100 ft (ADS-B data were recorded during the whole flight). The aircraft climbed to around FL 60 and continued east. When approaching the east border of the TMA, about 10 NM east of the airport, the aircraft made a turn and continued towards the southwest. In the middle of the TMA the aircraft made a wide 360-degree turn. When the test was repeated six weeks later at Arlanda the aircraft performed a similar flight at the same flight level as in the Kiruna test. These tests provided information about the accuracy of the ADS-B position at different distances from the ADS-B ground station inside the TMA.
After the flight tests were completed the log data was analysed and compared. The flights path reported via ADS-B was plotted in the same diagram as the flight path recorded from the RTK transceiver, see Figure 6-3. Then the deviations between ADS-B positions and RTK positions were statistically analysed and ADS-B maximum, minimum and mean deviation from the “true” RTK value was calculated. The results from these analyses are presented in Section 6.3.1.
67,2 67,3 67,4 67,5 67,6 67,7 67,8 67,9 68 68,1 68,2 68,3 18,5 19 19,5 20 20,5 21 21,5 22 22,5 Longitude (degrees) Latitude (degrees) Flight path Kiruna TMA Kiruna airport
Figure 6-1. Flight path for accuracy test inside Kiruna TMA.
6.2.1.2. Accuracy test in en-route airspace
The tests in en-route airspace in the Kiruna area (September 23) and in the Arlanda/Stockholm area (November 9) were performed close enough to the reference station for the RTK reports to have the best or second best accuracy, level 3 or level 2. The following data were recorded during the flight tests:
• GNSS data (aircraft)
• ADS-B data via VDL Mode 4 (ground station) • RTK data (aircraft)
• SSR data (from the radar station at Klöverträsk (about 100 NM south-southeast of Kiruna) and Arlanda radar station)
The logs were then compared to find out how much the ADS-B and the SSR positions differed from the true position, i.e. the position recorded in the RTK log.
The recording of RTK data started when the aircraft was about to leave the Kiruna TMA (see Figure 6-2), ADS-B and SSR data were recorded during the whole flight. Then the aircraft climbed to around FL 110 and continued southeast. About 40 NM southeast of Kiruna airport the aircraft made a turn and continued west
for approximately 50 NM. Finally the aircraft made another turn, northeast towards the airport. When re-entering the terminal area the test was finished. The same flight was repeated six weeks later but with Arlanda as the base instead of Kiruna. These tests provided information about the accuracy of the ADS-B position at different distances from the ADS-B ground station outside the TMA, in en-route airspace. The collected data was also used to compare ADS-B and SSR position accuracy. When the flight test was completed, the log data was analysed and compared. The flight paths reported via ADS-B and SSR were plotted in the same diagram as the flight path recorded from the RTK transceiver (see Figure 6-5). The deviations between ADS-B positions and RTK positions were then statistically analysed and ADS-B maximum, minimum and mean deviation from the “true” RTK value were calculated. The results from these analyses are presented in Section 6.3.2. Due to difficulties in the time synchronisation between the RTK log files and the SSR log files, the SSR data was never statistically analysed, so the only comparison made to the RTK logs was the plots.
67 67,2 67,4 67,6 67,8 68 68,2 68,4 18 18,5 19 19,5 20 20,5 21 21,5 22 22,5 23 Longitude (degrees) Latitude (degrees) Flight path Kiruna TMA Kiruna airport
Figure 6-2. Flight path for accuracy test in en-route airspace in the Kiruna area.
6.2.2. Method for verifying time stamp accuracy
To verify the time stamp accuracy the specifications from the manufacturer of the time reference system in the ground station have been used.
6.3. Validation of accuracy
To validate the accuracy performance of ADS-B/VDL Mode 4, log data from the tests described in Section 6.2 have analysed. Primary tools used in this work have been MS Excel and MS Access. The results from this analysis have then been compared with the requirements listed in Section 6.1 to see if ADS-B/VDL Mode 4 fulfils the requirements.
6.3.1. Validation of position accuracy inside terminal area
The technical requirements for SSR positional accuracy are somewhat difficult to transfer to ADS-B since the SSR requirements allow two types of error, range errors and azimuth errors. The operational requirement, a root-mean-square value of 300 metres from the true position, makes it easier to validate the ADS-B/VDL Mode 4 accuracy performance. For ADS-B there is only one type of error that is of interest, the deviation in metres from a reference point representing the true position.
With the reasonable assumption that the RTK position represents the true position, ADS-B position reports deviate from the true position inside the terminal area according to Table 6-2 below. The positions reported by the RTK and from ADS-B from the same tenth of a second have been compared.
Horizontal error in metres Kiruna test (no GRAS) Arlanda test (with GRAS)
< 25 m 38 % of the ADS-B reports 49 % of the ADS-B reports
< 50 m 60 % of the ADS-B reports 68 % of the ADS-B reports
< 100 m 80 % of the ADS-B reports 89 % of the ADS-B reports
< 200 m 96 % of the ADS-B reports 100 % of the ADS-B reports
< 300 m 100 % of the ADS-B reports
Table 6-2. ADS-B position error inside terminal area.
Conclusions that can be made from the values in Table 6-2 are that ADS-B fulfils the operational requirement (300 m) inside the terminal area without any problem. However, Table 6-2 also shows that GRAS does not enhance the accuracy performance as much as could be expected. In the other test at Arlanda airport, with stationary transceivers, GRAS enhanced the accuracy significantly, from a maximum deviation of about 140 metres without GRAS down to about 15 metres with GRAS, a big improvement, reducing the deviation to one tenth of the original deviation. But when analysing the flight test, where the (aircraft) transceiver has been in motion, GRAS does not improve the accuracy more than a few percent. Looking at plots from the flight shows that when the aircraft makes a turn is when the ADS-B track deviates the most from the RTK track. This is something that has to be examined further.
TMA test 67,55 67,6 67,65 67,7 67,75 67,8 67,85 20 20,1 20,2 20,3 20,4 20,5 20,6 20,7 20,8 20,9 21 Longitude (degrees) Latitude (degrees) RTK ADS-B Kiruna airport
Figure 6-3. Plotted ADS-B and RTK position reports inside Kiruna terminal area.
In Figure 6-3, the ADS-B position reports and the RTK position reports have been plotted next to each other. In the plot it is hard to separate one track from the other during most of the flight. However, in the 360º turn it is possible to see that ADS-B track deviates from the RTK track, although not by very much. This part of the plot has been enlarged and can be seen in Figure 6-4.
TMA test 67,58 67,59 67,6 67,61 67,62 67,63 67,64 67,65 67,66 67,67 67,68 67,69 67,7 20,35 20,4 20,45 20,5 20,55 20,6 20,65 20,7 20,75 Longitude (degrees) Latitude (degrees) RTK ADS-B
Figure 6-4. ADS-B and RTK position reports in the part of the flight where the tracks deviates the most.
6.3.2. Validation of position accuracy in en-route airspace
Of course the difficulties mentioned above to compare technical requirements for SSR accuracy remains in en-route airspace. The operational requirement is a root-mean-square value of 500 metres from the true position.
With the reasonable assumption that the RTK position represents the true position, ADS-B position reports deviate from the true position in en-route airspace according to Table 6-3 below. The position reported by the RTK and from ADS-B from the same tenth of a second has been compared.
Horizontal error in metres Kiruna test (no GRAS) Arlanda test (with GRAS)
< 25 m 34 % of the ADS-B reports 34 % of the ADS-B reports
< 50 m 46 % of the ADS-B reports 58 % of the ADS-B reports
< 100 m 96 % of the ADS-B reports 93 % of the ADS-B reports
< 200 m 100 % of the ADS-B reports 100 % of the ADS-B reports
Table 6-3. ADS-B position error in en-route airspace.
Conclusions that can be made from the values in Table 6-3 are that ADS-B fulfils the operational requirement (500 m) in en-route airspace without any problem. As with the position plot from the terminal area test (Figure 6-3) it is hard to see any difference between the RTK track and the ADS-B track. However, zooming in on a small part of the flight path shows that the ADS-B track deviates slightly from the RTK track also in en-route airspace, see Figure 6-5. The problem with GRAS not enhancing the position accuracy for moving targets, as discussed in Section 6.3.1, remains.
Figure 6-5. ADS-B, SSR and RTK position in a part of the flight where the three tracks clearly deviates.
6.3.3. Validation of altitude accuracy
Regarding the altitude accuracy there are two possible sources for altitude reports in ADS-B via VDL Mode 4; the aircraft pressure altimeter that gives the barometric altitude and the geometric altitude calculated by the GNSS. In aviation today the barometric altitude is used and this will probably be the case for years to come. The barometric altitude reported via VDL Mode 4 should not differ from the altitude reported via a SSR transceiver since the source is the same. Inside Kiruna terminal area there is no way to verify this since there is no radar coverage. However the analysis of the Arlanda log files verifies that there are no difference between the ADS-B altitude and SSR altitude.
In Figure 6-6 below the barometric altitude reported by ADS-B and SSR has been compared to the geometric altitude reported by the RTK equipment. The plot shows that during the test in Stockholm TMA the barometric altitude deviates about 500 ft, or 150 m, from the true altitude reported by the RTK. Beyond the scope of this document, but still, this shows that if ADS-B in a future case is to be used for approach and landing, the barometric altitude cannot be used. However, this is also the case with the geometric altitude in the ADS-B/VDL Mode 4 reports if GRAS is not available, as shown in Table 6-4 and Table 6-5.
En-route test Kiruna (part of)
1 km 67,07 67,08 67,09 67,1 67,11 67,12 67,13 67,14 67,15 21,35 21,4 21,45 21,5 21,55 21,6 Longitude (degrees) L a titu d e (d eg rees) RTK ADS-B SSR
Altitude accuracy in TMA 5000 5200 5400 5600 5800 6000 6200 6400 90165800 90165900 90166000 90166100 90166200 90166300 90166400 90166500 90166600 90166700
Time in seconds since 2000-01-01
Altitude in feet
SSR (barometric alt) ADS-B (barometric alt) RTK (geometric alt)
. Figure 6-6. SSR and ADS-B barometric altitude vs RTK geometric altitude.
The geometric altitude in the ADS-B reports can be compared to the geometric altitude in the reports from the RTK equipment. Table 6-4 and Table 6-5 contains the result from these comparisons. Note that differential corrections of the GPS position (GRAS) was not available during the Kiruna test and therefore the altitude deviation may be greater than expected.
Kiruna test (no GRAS)
Minimum error: 0 metres
Maximum error: 86 metres
Mean error: 28 metres
Standard deviation from mean value: 38 metres
Table 6-4. Geometric altitude error inside terminal area
Kiruna test (no GRAS)
Minimum error: 0 metres
Maximum error: 144 metres
Mean error: 42 metres
Standard deviation from mean value: 56 metres
Table 6-5. Geometric altitude error in en-route airspace.
6.3.4. Validation of time stamp accuracy
According to the manufacturer of the time reference system (Thales Air Traffic Management) the accuracy is UTC +/- 50 nanoseconds. To verify this is subject to further work.
7. Availability
7.1. Availability requirements
7.1.1. SSR availability requirements
In Table 7-1 below the availability requirements for SSR data can be found ([Ref 23] Section 5.3.3). A radar sensor shall be considered unavailable if no radar target reports, including field monitor(s), are produced for more than two antenna scans. The requirements below apply to the overall availability. The availability of an individual sensor may be lower provided that other radar stations contribute to satisfy the overall requirements.
System Status Availability
(Equivalent Annual Outage Time) Maximum Instantaneous Outage Time
Full Data / Full Performance 0.995
(44 h)
4 h Full Data / Reduced Performance 0.999
(9 h) 10 min
Essential Data 0.99999
(6 min) 10 s
Table 7-1. Radar Surveillance Data Availability.1
Outage time means the time that the system is not available. Full data are:
• Aircraft horizontal position and history • Aircraft identification
• Aircraft vertical position
• Specific indication Mode A special codes (i.e. 7500,7600,7700) • Ground speed
• Status of the track whether it is primary, secondary, combined or extrapolated. Essential data are:
• Aircraft horizontal position and history • Aircraft identification or Mode A code; • Aircraft vertical position
Full performance means that all elements and functions of the radar chain are operating normally to the performance of this standard. Reduced performance means that the performance of some element of the radar chain is below full performance.
7.1.2. ADS-B availability requirements
The requirements for ADS-B data availability as stated by ICAO ([Ref 15] Part I, chapter 3, appendix A), see Table 7-2, almost equals the requirements for SSR essential data. The content of an ADS-B message however correspond to full data for a SSR message.
1 Scheduled maintenance, scheduled non-operational hours and force majeure events excluded
Definition
Availability is the probability that a system can provide the required function and performance within the specified service volume at the start of an intended operation [Ref 26].
Availability
(Equivalent Annual Outage Time)
Maximum Instantaneous Outage Time 0.99996
(21 min) N/A
Table 7-2. ADS-B data availability requirements.2
Note: The EUROCONTROL availability requirement is 0.9995, in other words lower than the ICAO
requirement.
7.2. Verification of availability
To test the availability log files from the flight tests in Kiruna and at Arlanda airport have been analysed, but if the validation is going to be statistically significant more log data has to be analysed. Therefore two stationary transceivers at Arlanda airport have been transmitting ADS-B data during a longer period of time, the last two weeks in November 2002. During two weeks, with an update rate of one report per second, approximately 2.5 million ADS-B reports are generated. The location of the transceivers can be seen in Table 7-3 and Figure 7-1. It is not obvious that the results from this test at the airport ground level are applicable to terminal area and en-route airspace. Nevertheless it could give a hint about availability performance. For further information about the static transceiver test at Arlanda airport, see appendix A.3.
Figure 7-1. Static transceiver test, Arlanda.
2
Transceiver Latitude Longitude
Ground station 59.653230 17.916155
Transceiver 1 59.649652 17.937530
Transceiver 2 59.637176 17.957397
Table 7-3. Location of the transceivers used in static test at Arlanda airport.
As a criteria to see if the system has been available or not, the SSR availability criteria has been used (Section 7.1.1). That is if more than two consecutive reports are missing, the system is considered to be unavailable. In the test from Kiruna and Arlanda an update rate of one report per second has been used, i.e. if the time difference between two consecutive reports is more than three seconds, the system has not been available.
Tools that have been used in the availability analysis are Matlab, MS Access and MS Excel. For calculation methods and definitions, see Appendix C.3.
7.3. Validation of availability
The log data from Kiruna contains data from approximately three hours and 25 minutes of flight, or 12319 seconds to be precise. 12155 reports can be found in the log file, i.e. 164 reports are missing and 98.67 % of the reports have been successfully transmitted. However, there are never two or more consecutive reports missing and hence the availability has been 100 % during the Kiruna flight test. The log data from the Arlanda flight test contains data from approximately five and half hour of flight, or 19972 seconds. 19351 reports can be found in the log file, i.e. 96.89 % of the report has been successfully transmitted. At one occasion there are more than two consecutive reports missing which means that the availability in the Arlanda flight test has been 0.999845 or 99.9845 %. Adding the reports from the two flight-tests the availability has been 0.99990. This fulfils the EUROCONTROL availability requirement for ADS-B, which is 0.9995, but it is a bit lower than the ICAO ADS-B requirement at 0.99996. More data to evaluate is needed to verify these results and to make them statistically significant.
Availability (%)
Kiruna flight test 100.00
Arlanda flight test 99.98
Table 7-4. Availability in flight test in Kiruna and at Arlanda.
The availability of the stationary transceivers at Arlanda airport has not been quite as high as the availability from the flight test. The result can be seen in Table 7-5 below.
Availability
Transceiver 1 (%) Transceiver 2 (%) Availability
Week 1 99.97 98.01
Week 2 99.83 99.96
Table 7-5. Availability in stationary transceiver test at Arlanda airport.
The difference between the flight tests and the stationary test is so small that it is possible that with more log data to evaluate, the results could have been more similar, or the differences could have been larger. More data to evaluate is needed to verify these results. Possible explanations to the somewhat lower availability at the airport could be the background noise level at the airport or buildings blocking the line-of-sight. Another possible cause could be that at Arlanda the frequency next to one of the VDL Mode 4 frequencies is used by another service, the Aircraft Communications Addressing and Reporting System (ACARS). This will not be the case when ADS-B/VDL Mode 4 becomes operational.
