Assessment of VDL Mode 4 Frequency, Capacity and Performances

Transcription

1 EUROPEAN ORGANISATION FOR THE SAFETY OF AIR NAVIGATION E U R O C O N T R O L Assessment of VDL Mode 4 Frequency, Capacity and Performances TRS041 Deliverable 2.1: Contribution to Frequency Planning Criteria Definition & Airborne Co-Site Interference Assessment (Update 2009 included) Edition : V4.1 Edition Date : 13th October 2010 Status : Release Issue Class :

2 DOCUMENT IDENTIFICATION SHEET DOCUMENT DESCRIPTION TITLE Assessment of VDL Mode 4 Frequency, Capacity and Performances EWP DELIVERABLE REFERENCE NUMBER PROGRAMME REFERENCE INDEX EDITION: V4.1 Abstract EDITION DATE: 13th October 2010 This document originally was the second deliverable of contract C/1.189/00/EC/TRS041/05. It has been updated in 2009 for the extension of analysis for the airborne co-site assessment. DOCUMENT STATUS AND TYPE STATUS CATEGORY CLASSIFICATION Working Draft Executive Task General Public Draft Specialist Task EATM Proposed Issue Lower Layer Task Restricted Released Issue ELECTRONIC BACKUP INTERNAL REFERENCE NAME: HOST SYSTEM MEDIA SOFTWARE(S) Microsoft Windows Type: Hard disk Media Identification:

3 DOCUMENT CHANGE RECORD The following table records the complete history of the successive editions of the present document. EDITION DATE REASON FOR CHANGE SECTIONS PAGES AFFECTED 0.1 Dec. 25 th 2005 Creation All 0.2 February 2006 Update with laboratory results and analysis All March- April Updates after co-site tests analysis and All 2006 specific simulations addressing co-site impact nd May 2006 Editorial changes and text refinements All 2.0 July 2009 Update for the airborne co-site assessment. All 3.0 Dec Updated results taking into account proposed change in recommended VDL Mode 4 set of system parameters (PS4). All, but sections 1 & Sept 2010 Minor editorial changes Section Oct Minor editorial changes Section 1.1 Section Edition: V4.1 Page 3

4 TABLE OF CONTENTS 1. INTRODUCTION BACKGROUND ON VDL 4 ANALYSIS AT EUROCONTROL VHF Interference investigations System capacity studies through simulations STUDY OBJECTIVES PRESENT DELIVERABLE OBJECTIVES DOCUMENT STRUCTURE VDL MODE 4 FREQUENCY PLANNING CRITERIA INTRODUCTION Background on ICAO ACP work for definition of separation criteria Background on the interference testing methodology developed for ICAO ACP FREQUENCY PLANNING RESULTS DSB-AM and VDL Mode VDL Mode 2 and VDL Mode VDL Mode 4 versus another VDL Mode 4 channel CONCLUSION AND RECOMMENDATION AIRBORNE CO-SITE INTERFERENCE INTRODUCTION ANALYSIS IN LABS OF VDL 4 INTERFERENCE OVER VHF AM-DSB VOICE TRAFFIC ANALYSIS OF COSITE AM-DSB VOICE INTERFERENCE OVER VDL 4 RECEPTION Introduction Co-site interference scenarios and results definition Applied to CPDLC and time-critical applications Applied to ADS-B applications Applying measured rejection performance in simulations CONCLUSIONS OF PRESENT DELIVERABLE Edition: V4.1 Page 4

5 1. INTRODUCTION 1.1 Background on VDL 4 analysis at Eurocontrol EUROCONTROL has carried out several studies about air/ground communications subnetworks covering interference, frequency planning and capacity aspects for several of them. Two main VHF data-link sub-networks have been studied in depth: VDL Mode 2, that currently is deployed to replace POA ( Plain Old ACARS ) and to support CPDLC services in the framework of Link2000+ program; and VDL Mode 4, promoted as a complementary candidate to VDL Mode 2 for point-to-point applications (also candidate for supporting ADS-B as an alternative to Mode S) VHF Interference investigations Frequency planning criteria VDL Mode 2 Frequency planning criteria were developed in ICAO Working Group B between 1998 and 2001 and were accepted in September 2001 in agreement with the main contributions made by EUROCONTROL [10]. VDL Mode 4 Frequency planning criteria development has been initiated in the same period in ICAO ACP WG-B. The group opted for the same methodology [15] and defined interference assessment criteria adequate for VDL Mode 4. Unfortunately, the first testing campaign in 2002 revealed reception performance problem on the single VDL Mode 4 airborne radio available for these tests. Therefore assessing a realistic frequency planning based on these measurements had not been possible. The same manufacturer s modified airborne equipment has been used in the new testing campaign in March 2005 again. However the overall results did not differ much from those obtained in The critical test case that had been observed in 2002 has been observed again; it appeared that this was due to a premature saturation of the receiver. A repetition of the same tests with a VDL4 ground radio from a different manufacturer yielded better results roughly comparable with those achieved in the VDL2 tests. The saturation problem has been identified and resolved in a new airborne equipment release in September Prior to testing that radio again, contacts had been taken with several avionics radio manufacturers but in December 2005, none additional avionics had been made available to EUROCONTROL. 1 1 The avionics radio tested is a product of CNS. ADSI, RTX and Rockwell Collins were contacted in first semester of Edition: V4.1 Page 5

6 Airborne Co-Site Interferences EUROCONTROL has completed several studies addressing interference issue in the VHF band over the last few years. When frequency planning criteria has been established for VDL Mode 2 [10], ICAO ACP WG-B has requested Airborne co-site interferences to and from VDL Mode 2 also to be analysed [11]. The same crucial subject had been addressed also when frequency planning criteria development has been started for VDL 4 within ICAO WG-B [15]. Airborne co-site interference case is the most stringent co-site case to be addressed when evaluating VDL compatibility with other existing VHF systems (within MHz). As a matter of fact, antenna isolations are poor due to the short antenna separation distances on the aircraft, and transmission powers in use need to be high enough to provide a wide range of operation, although it potentially conflicts with the uplink VHF voice or data signals to be correctly and independently decoded. In 2003, a VDL Mode 4 Airborne Architecture Study [5] has analysed the theoretical impact of VHF co-site interference in general and showed that detailed and practical investigations were necessary to proceed further. NUP II-project has operated ADS-B trials based on VDL 4 for several years and also identified a few times in trials the phenomenon of airborne co-site interference, resulting in an issue listed and to be properly addressed. [13] VDL Mode 4 is foreseen to support a large field of applications including ADS-B and point-to-point communications, and is intended also for supporting so called timecritical applications. In support of these applications, investigations about interference to VDL Mode 4 and their impact on VDL 4 capabilities become crucial. Other VHF systems currently used in the air traffic context need to be protected against the potential interference of VDL Mode 4. It is the case of DSB-AM voice which will remain safety critical for long, but also VDL 2 which is the first data-link generation able of fulfilling other ATC requirements; one scenario could be the coexistence of VDL Mode 2 and Mode 4 on the same aircraft. The first phase of VDL 4 Co-Site interference Impact Investigation including airborne co site scenario and detailed test description has been achieved in 2004 by ISA Telecoms on behalf of EUROCONTROL [8], [9] System capacity studies through simulations Previous VDL4 simulation campaigns have been conducted by Helios Technology on behalf of EUROCONTROL. The first campaign has been conducted in 2002 to address the VDL4 capacity in support of ADS-B broadcast applications. The second campaign has been conducted, completing in 2005 [6] to address the VDL4 capacity in support to point-to-point communications. Referring to criteria of CoopATS [22] and Macondo [23] available at time, the campaign demonstrated that, as currently specified in ICAO standards [1], the VDL4 technology was unsuitable to support time-critical and/or safety critical applications. Having identified some of the protocol issues, the campaign investigated potential protocol enhancements [6]. Edition: V4.1 Page 6

7 Changes were recommended by Helios Technology in several potential directions (optimisation of re-transmission parameters, ground-coordination development, sectored antennas,..). Although some capacity improvements were expected, the identified enhancements were still requiring additional validation before proceeding with submission to ICAO for endorsement. End 2005, EUROCONTROL contracted ALTYS Technologies for additional simulations, both to confirm previous results, and to proceed further with the evaluation of the technology if it was to be deployed and used in Europe in support to ATC communications. A first set of conclusions were presented to the community in a workshop help May 2006 in EUROCONTROL premises. Following the workshop, and to further investigate potential enhancements, EUROCONTROL pursue the analysis, coordinating with VM4C members about most desirable and acceptable enhancements. 1.2 Study Objectives The objectives of the present VDL Mode 4 study are as follows: Pursue and complete co-site interferences investigations initiated in 2003 Complete the VDL Mode 4 interference testing to contribute to ICAO ACP WG B on planning criteria Validate and complement previous system capacity assessment campaigns in support of time-critical and safety-critical applications 2. To achieve the above, the study has been structured around three Work Packages: WP1 Project Management WP2 Interferences investigations with a deliverable D2 reporting on Frequency Planning Criteria and Airborne Cosite interference assessment. WP3 Simulators Cross-check and Capacity Assessment by use of Simulations, with a deliverable D3 reporting on VDL 4 capacity The objectives of latest VDL Mode 4 study (up to 2009) were, in coordination with VM4C, to further investigate and assess VM4C identified enhancements. In practice, recommended changes dealt with changes in VDL4 System Parameters, to supersede default values defined in ICAO SARPS (cf. [1]). Additional activities have thus been conducted: Validation of ACTS/VDL4 physical model through theoretical and cross-check simulations; Reassessment of co-site interferences (see updated D2: present document). Rerun of capacity simulations based on VM4C provided parameter set (see updated D3; section 4.3). 2 Although such exercise might be valuable as well, the present study has no objectives with respect to the validation of previous system capacity assessment campaigns in support of broadcast applications. Edition: V4.1 Page 7

8 1.3 Present Deliverable Objectives The present deliverable documents the results and findings of the VDL4 interferences investigations study (WP2) of the above mentioned study. The main objectives of WP2 are: Achieve testing and analysis of interference involving VDL Mode 4 in view of contribution to ICAO ACP WG-B on the definition of frequency planning criteria. ICAO initiated some years ago VDL 4 criteria development within being able to conclude it so far. Complete the Airborne Co-Site investigations initiated in a previous EUROCONTROL TRS (063-04) and check their results compatibility with the frequency planning criteria results. Version 3.0 (dated 2009) of present deliverable also includes an update of all results, based on VDL Mode 4 set of system parameter (also referred to as PS4 ) recommended 3 by LFV in It is the synthesis of such complementary assessments, all conducted between up to mid Document Structure Section 2 develops a proposal for VDL Mode 4 frequency planning criteria, based on the analysis of relevant measurement made during the testing campaigns performed between 2002 and end 2005, unfortunately based on a single VDL 4 avionics radio, and one ground-station.: EUROCONTROL Brétigny lab and AIR FRANCE Industries in 2002 [15] EUROCONTROL Brétigny lab in March EUROCONTROL Brétigny lab and AIR FRANCE Industries in December Sub section 2.2 gives analysis of the main results and recommendation for criteria. Section 3 investigates Airborne Co-site interferences involving VDL Mode 4. This section has been extended in 2009 with the analysis made with a new set of VDL Mode 4 system parameters (detailed in D3, Appendix C.5). Appendix A and Appendix B gives index and abbreviations. Appendix C is recalling the main causes of interferences. Appendix D is recalling the frequency planning methodology. Appendix E provides tests parameters. Appendix F gives the detailed frequency planning results. Appendix G provides details about DSB-AM interfered by VDL Mode 4 recordings. Appendix H recalls the VDL Mode 4 assessment methodology 3 Recommended Parameter Set PS4 is fully documented in D3 Edition 2.0 (or latest), Appendix C.5. Edition: V4.1 Page 8

9 Appendix I provides the detailed results of co site measurement Appendix J provides the co site test set up. Edition: V4.1 Page 9

10 2. VDL MODE 4 FREQUENCY PLANNING CRITERIA 2.1 Introduction Background on ICAO ACP work for definition of separation criteria The present section presents EUROCONTROL testing results for VDL Mode 4 frequency planning criteria based on ICAO ACP WG-B methodology. It provides material for a contribution to ICAO ACP in view of defining the separation rules (amount of guard-band channels between VDL Mode 4 and other VHF systems, VDL Mode 2 and DSB-AM in this case) necessary to avoid unacceptable interference occurrences in operational conditions. EUROCONTROL conclusions are derived from several testing campaign. For each campaign, only the relevant set of results has been kept Background on the interference testing methodology developed for ICAO ACP It must be recalled that interference impacts depend both on interferer (wide-band channel emissions) and on victim receiver characteristics (rejection performance). Furthermore, no requirements had been put in the ICAO standards for the DSB-AM radios transmitter spectrum (there is no spectrum mask comparable to those of VDR specifications for VDL modes). In addition, many radios have better performances than required by the standards. Therefore, the method is intended to spare any possible spectrum by measuring a representative set of production standard compliant avionics and ground radios, both for interferer and victim sides, in order to define the separations really required. Also the method is made of objective and reproducible tests for several organisations and labs to be able to contribute to ICAO ACP frequency planning criteria taking into account commonly agreed the worst operational cases to be protected. The present EUROCONTROL test program thus consisted in testing VDL Mode 4 vs. VDL Mode 2, DSB-AM and a second VDL Mode 4 channel. The test methods that have applied at EEC-Bretigny and Air France Industries labs are summarised in table 1 below and results presented in the next section 2.2: Victim Vs. Interference Source DSB-AM VDL 2 VDL 4 DSB-AM - - Method 2 VDL Method 3 VDL 4 Method 3bis Method 3bis Method 3bis Edition: V4.1 Page 10

11 Table 1 - ICAO Test methods used for VDL4 interference testing The following operational scenarios have to be tested Scenario 1 Aircraft on the Ground vs. Aircraft on the Ground Two aircraft situated at adjacent gates communicating on different frequencies with an assumed physical separation of 210m when the minimum signal level is 82 dbm. Scenario 2 Aircraft on the Ground vs. Ground Station One aircraft on gate communicating on one frequency and a ground station communicating with another aircraft where it is assumed that the minimum physical separation between antennas is 750m when the minimum signal level is 93 dbm. Scenario 4 Aircraft in Flight vs. Aircraft in Flight Two aircraft in flight flying parallel tracks communicating on two different frequencies with an assumed vertical physical separation of 600 meters (2,000ft) between the antenna. 2.2 Frequency Planning Results This section summarises the dimensioning interference results and provides the EUROCONTROL recommendations for a realistic and safe planning criteria between VDL Mode 4 and other aeronautical VHF systems. According to ICAO ACP method, the figures representing the minimum separation distance between the victim receiver and interferer transmitter are made with the worst results collected during different testing campaigns. Tests are achieved for three frequencies (low band 118MHz, middle band 128MHz and upper band 136MHz). The worst case of the 3 frequencies has been represented with a dash bold grey curve. Detailed results from which the dimensioning results are derived are given in Appendix F (including Desired /Undesired (D/U) curves) about DSB-AM and VDL Mode 4 separation. Edition: V4.1 Page 11

12 2.2.1 DSB-AM and VDL Mode DSB-AM to VDL Mode 4 interferences DSB-AM To VDL Mode 4 Worst Minimum Separation Distance Vs Adjacent Channel # 136 MHz Worst Case 128 MHz Worst Case 119 MHz Worst Case All Frequencies Worst m Air Scenario Distance (m) m Ground Scenario Adjacent channel # Figure 1 - DSB-AM to VDL 4 Worst case Minimum separation distance Air frequency planning scenario (Scenario 4) The required protection distance in scenario 4 is 600 m. Observation of Figure 1 - shows that this distance is reached at the 2nd adjacent channel for the 2% MER criteria. Therefore 1 guard band channel is sufficient to protect a VDL Mode 4 channel from a DSB-AM channel with MER superior or equal to 2% when the air frequency planning scenario is considered Ground frequency planning scenario (scenario 1 & 2) It has been proven that, because of special protection techniques available for ground station [12], the dimensioning scenario (worst case to be considered) was scenario 1 (Aircraft on the Ground vs Aircraft on the Ground) The required protection distance in scenario 1 is 210 m. Observation of Figure 1 - shows that this distance is reached at the 5th adjacent channel any test configuration. Therefore when scenario 1 is considered, 4 guard-band channels are recommended to protect a VDL Mode 4 channel from a DSB-AM channel. Edition: V4.1 Page 12

13 VDL Mode 4 to DSB-AM interferences It has been proven that the most stringent interference criteria is Signal/Pulse (S/P) >12dB. Squelch lifting is not the dimensioning criteria VDL Mode 4 To DSB-AM Worst Minimum Separation Distance Vs Adjacent Channel # 136 MHz Worst Case 128 MHz Worst Case 119 MHz Worst Case All Frequencies Worst m Air Scenario Distance (m) m Ground Scenario Adjacent channel # Figure 2 - VDL4 to DSB-AM Worst case Minimum separation distance Air frequency planning scenario (Scenario 4) The required protection distance in scenario 4 is 600 m. Observation of Figure 2 - shows that this distance is reached at the 4th adjacent channel for the 12dB S/P criteria. Therefore 3 guard-band channels are sufficient to protect a DSB-AM channel from a VDL mode 4 channel with S/P equal to 12 db when the air frequency planning scenario is considered Ground frequency planning scenario (Scenario 1 and 2) The required protection distance in scenario 1 is 210 m. Observation of Figure 2 - shows that this distance is reached at the 5th adjacent channel any test configuration. Therefore when scenario 1 is considered, 4 guard-band channels are recommended to protect a DSB-AM channel from a VDL mode 4 channel. Edition: V4.1 Page 13

14 Conclusion on DSB-AM Taking into account the results of and , and based on the set of one avionics and one ground radio tested, it is required to have: 4 guard-band channels between a VDL Mode 4 and a DSB-AM channel in the same service volume when the ground scenario is considered. 3 guard-band channels between a VDL Mode 4 and a DSB-AM channel in the same service volume when the air scenario is considered VDL Mode 2 and VDL Mode VDL Mode 4 to VDL Mode 2 interferences The only channel load configuration considered for frequency planning is made of 6.67% VDL Mode 4 interferer load and 20% VDL Mode 2 victim load as per [15] and [17]. However it can be observed that even with 100% VDL Mode 4 interfering channel load, the requirements are almost fulfil in the following scenarios. The full set of results are presented in Appendix F VDL Mode 4 To VDL Mode 2 Airborne -82dBm Worst Minimum Separation Distance Vs Adjacent Channel # 136 MHz 128 MHz 119 MHz 210m m Scenario Distance (m) m Scenario 210m Scenario Adjacent channel # Figure 3 - VDL 4 to VDL2 victim Airborne at -82dBm Edition: V4.1 Page 14

15 VDL Mode 4 To VDL2 Ground Station -93dBm Worst Minimum Separation Distance Vs Adjacent Channel # 136 MHz 128 MHz 119 MHz Distance (m) Adjacent channel # Figure 4 - VDL 4 to VDL2 victim Ground station at -93dBm Air frequency planning scenario (Scenario 4) The required protection distance in scenario 4 is 600 m. Observation of Figure 3 - shows that this distance is reached at the 2nd adjacent channel. Therefore a single guard-band channel is required to protect a VDL Mode 2 channel from a VDL Mode 4 channel in the same service volume when the air scenario is considered Ground frequency planning scenario (Scenario 1 and 2) The required protection distance in scenario 1 is 210 m. Observation of Figure 3 - shows that this distance is reached at the 2nd adjacent channel. The required protection distance in scenario 2 is 750m. Figure 4 - shows that this distance is easily reached from the 2 nd adjacent channel on. As stated earlier specific filtering techniques can still improve Ground stations interference immunity performance. Therefore a single guard-band channel is required to protect a VDL Mode 2 channel from a VDL Mode 4 channel in the same service volume when the ground scenarios are considered VDL Mode 2 to VDL Mode 4 interferences The channel load configuration of the presented results is 2% interfering VDL Mode 2 and 100% victim VDL Mode 4. Edition: V4.1 Page 15

16 VDL Mode 2 (2%) To VDL Mode 4 Worst Minimum Separation Distance Vs Adjacent Channel # 136 MHz Worst Case 128 MHz Worst Case 119 MHz Worst Case All Frequencies Worst Case m Air Scenario Distance (m) m Ground Scenario Adjacent channel # Figure 5 - VDL 2 to VDL4 Worst case Minimum separation distance Air frequency planning scenario (Scenario 4) The required protection distance is 600 m. Figure 5 - shows that this distance is reached at the 2nd adjacent channel. Therefore a single guard-band channel is required to protect a VDL Mode 4 channel from a VDL Mode 2 channel in the same service volume when the air scenario is considered Ground frequency planning scenario (Scenario 1 and 2) The required protection distance in scenario 1 is 210 m. Observation of Figure 5 - shows that this distance is reached at the 2nd adjacent channel. Therefore a single guard-band channel is required to protect a VDL Mode 4 channel from a VDL Mode 2 channel in the same service volume when the ground scenarios are considered Conclusion on VDL Mode 2 Taking into account the results of and , it is recommended to have: 1 guard-band channels between a VDL Mode 2 and a VDL Mode 4 channel in the same service volume. Edition: V4.1 Page 16

17 2.2.3 VDL Mode 4 versus another VDL Mode 4 channel No new measurement has been achieved during the last campaign in this configuration. However [15] was recommending 1 guard-band channel between two VDL Mode 4 channels in the same service volume. Taking in to account that the last avionic modifications have increased the receiver immunity performance, it is likely to consider that the recommended criteria is still appropriate with the new hardware. 1 guard-band channel between two VDL Mode 4 channels in the same service volume for any scenario. 2.3 Conclusion and recommendation The following table summarises the VDL Mode 4 frequency planning criteria recommendations, based on the testing campaign achieved on one VDL Mode 4 avionics and one ground radio: System Table 2 - Recommended amount of (25 khz) Guard-band channels for allocation of a VDL Mode 4 channel DSB-AM 4 in ground scenario 3 in Air Scenario VDL 2 1 VDL 4 1 Contribution to VDL Mode 4 frequency planning criteria : summary Those results should be presented to ICAO ACP for consideration and compilation with other independent laboratories results applying the same method on more avionics and ground radios. Also, ICAO ACP process is planning to consider sufficient airborne co-site analysis material (subject of the next section in the present document) to check compatibility with draft frequency planning criteria before conclusion into definite VDL 4 frequency separation rules. Edition: V4.1 Page 17

18 3. AIRBORNE CO-SITE INTERFERENCE 3.1 Introduction Airborne co-site interference is the undesirable effect caused on a received signal (the victim) by a concurrent transmission (the interferer) sent onboard the same aircraft. Typical cases of airborne co-site interference in the VHF aeronautical band are the interference made: By onboard data transmissions (ACARS, VDL Mode 2 or 4) over onboard voice reception (especially critical for Air Traffic Control Communications) By the pilots voice transmissions (for ATC and OPC communications purpose) over air-to air or ground-to-air data reception onboard. Interferer signal Victim signal (from either air or ground) Figure 6 - Airborne Co-site Interference cases The present sections analyses, through laboratory measurements and through simulations for a part, the impact of airborne co-site interference involving VDL Mode 4 avionics as interferer and as victim. The content delivers : Assessment of VDL Mode 4 co-site interference on ATC voice communications in worst and in typical operational scenarios. Measure the impact of onboard AM-DSB co-site interference on VDL Mode 4 operations including point to point and ADS-B. Edition: V4.1 Page 18

19 3.2 Analysis in labs of VDL 4 interference over VHF AM-DSB voice traffic The tests produced audio records achieved in Eurocontrol experimental centre in Bretigny. The worst cases to be analysed are the cases where: Frequency separation between VDL and victim voice signal is the minimum separation likely to become operationally possible VDL-to-voice antenna isolations are covering the smallest encountered values (e.g. for some small aircraft : 35 db + 3 db for cables and connectors) The victim voice signal is the theoretical minimum required by ICAO (-82 dbm for avionics), equivalent to the one received by an aircraft flying at the boundary of a service area. Ideally tests should at least be done in accordance with the worst case VDL 4- downlink loading i.e. the most demanding point-to-point and ADS-B applications. In the present case, the avionics tested was not able to transmit more than one time per second. As criteria for assessment by operators of the interference on voices, absence of squelch opening, severity and frequency of interference (the audible clicks ) heard over the desired signal are likely to be considered by ICAO ACP. Samples of interference have been recorded for different combinations of message transmissions rates, frequency separation, and antenna separation. Those are listed in [Appendix G]. The reader is invited to listen to typical samples 4. Also he could try to figure for himself what pilots representative would tolerate or not. As a preliminary feedback, it seems that: Based on the present test installation, the sample for one of the worst case (ADS B report sent every second, on the 2 nd adjacent channel with a total interferer to victim signal isolation of 40 db ), seems to be too heavily interfering voice. At the opposite, the favourable case of ADS-B reports sent every 5 sec, with 1 MHz of separation and 50 db of isolation, seems a tolerable interference in practise. Squelch opening has been tested at 40 db, with VDL 4 transmission every 5 sec in the 2 nd adjacent channel and for 1-slot and 2 slots transmission cases. No occurrence of squelch opening was observed as expected with one or two slots transmissions. It was not possible to test more multi-slot cases. 4 Those samples should be made available at ICAO ACP meetings, and other VDL 4-workshops, and it is intended to make them also available on Eurocontrol web site at the time that the present material will be published. Edition: V4.1 Page 19

20 3.3 Analysis of cosite AM-DSB voice interference over VDL 4 reception Introduction In , EUROCONTROL developed an extensive study aiming at developing a complete methodology and test plan to assess the impact of voice interference on all kinds of VDL Mode 4 applications. (Point-to-point, ADS-B, ADS-C ). Dimensioning scenarios involving different aircraft parameters (Transmitting power antenna coupling), frequency separation, background traffic levels, and possible VDL mode 4 usages have been published. Please refer to ([8], [9]) for a complete view of the method and developed scenarios. The testing campaign that took place in AIR FRANCE INDUSTRIES and EUROCONTROL EEC during December 2005 was too short to achieve the initial exhaustive scenarios list. Therefore it has been decided to pay a particular attention to commercial aircraft configuration. Furthermore most of the point to point scenarios have been constructed originally on the basis of default VDL Mode 4 standard parameters. However, if the default parameter set would be changed in the standards, new calculations could be done without need to repeat the testing. This has been applied in A new set of VDL Mode 4 parameters (also referred to as Parameter Set PS4, detailed in D3, Appendix C.5) have been proposed and a final evaluation of the airborne co-site interference has been achieved (present document, Edition 3.0 and above). Since the current most important ATC Data link applications are CPDLC and ADS- B, the analysis focuses on their potential quality of service degradation in presence of an on-board voice interferer. Simulations with and without interference have also been ran to estimate the contribution part of co-site interference on the overall system performance Co-site interference scenarios and results definition The results are presented in the following way: For each scenario, the maximum protected distance (MPD) between the desired transmitter and the receiver in presence of interference is given in NM on the basis of specified QOS criteria. In other words, any VDL 4 station transmitting station located within the MPD will be received fulfilling the QOS criteria. The criteria will not be fulfilled anymore for stations outside this area. Since the distance calculation is based on free space propagation model and therefore biased to some extent, the results granularity used is 5 NM. That explains why they can be only <5NM, 5NM, 10NM Edition: V4.1 Page 20

21 Maximum Protected Distance (MPD) Co site victim Aircraft Figure 7 - Maximum Protected distance (MPD) area The scenarios carried out apply for a commercial aircraft, with the minimum coupling between antennas guaranteed by the main airframes Airbus and Boeing. These figures include 3dB feeder losses: 53 db for opposite sides antennas 38 db for same side antennas It has to be noted that other type of aircraft initially considered such as general aviation are even more constraining. The victim channel frequencies under test were 119, 128 and 136 MHz. The 5 th (125 khz), 20 th (500 khz), and 40 th (1 MHz) adjacent channels have been chosen for the interferer to victim frequency separation Applied to CPDLC and time-critical applications This paragraph thus presents a sample of the method developed in the previous study ([8],[9]) and applied on a data stream like those of point-to-point exchanges. The performance requirements introduced by CoopATS and MACONDO and the Roadmap data link study for time-critical applications are recalled here: Time Criticality Message Category 95% Time Delay (One way) 99,996% Time Delay (One way) Critical C Critical Currently undefined Very V Distress High indicating grave and imminent danger 1 s 3 s 2 s 5 s Edition: V4.1 Page 21

22 Time Criticality Message Category 95% Time Delay (One way) 99,996% Time Delay (One way) High H Urgent having a potential impact on the safety of the aircraft or persons onboard or within sight Mediu m M Flight Safety Comprising movement and control messages and meteorological or other advice of immediate concern to an aircraft in flight or about to depart, or of immediate concern to units involved in the operational control of an aircraft in flight or about to depart 5 s 15 s 10 s 20 s Low L Routine Surveillance or Navigation 30 s 60 s Table 3 - Maximum End to End One way Transmission Delay The third column indicates target delay value to be respected by 95% of the successful one-way transfers. The fourth column gives target delay to be respected by 99,996% of successful one-way transfers. Those figures are applicable for endto-end communications, i.e. from the sender side, up to the receiver side, including the processing time required to propagate the information through the ground infrastructure or display it on a unit (pilot or controller) whenever applicable. The critical category has been left undefined, and currently is not further considered. The present intention is to check what is feasible on VDL Mode 4 of the V, H, M, and L categories taking the airborne co-site interference into account. Due to the bottleneck that constitutes air-ground communications, the assumption is that 75% of the allowed time shall be allocated to the Data Link transmission, which leads to the following table: Category number Message Category Criteria (a) 95% Criteria (b) 99,996% 1 Distress 1,50 sec (1a) 3,75 sec (1b) 2 Urgent 3,75 sec (2a) 11,25 sec (2b) 3 Flight Safety 7,5 sec (3a) 15 sec (3b) 4 Routine Surveillance or Navigation 22,5 sec (4a) 45 sec (4b) Table 4 - Maximum One way -VDL Transmission Delay In Appendix I.3 a data profile is recalled, that was created in ([8], [9]) and is corresponding to the traffic induced by all ATC and AOC data applications. The probability of collision of each data message received with an airborne co-site voice message is reported, for each flight phase. From there, for different VDL 4 traffic levels on the channel, the maximum MER (Message Error Rate) tolerable is Edition: V4.1 Page 22

23 deduced, together with the maximum distance between the data source and the aircraft of interest Large aircraft, default VDL Mode 4 parameters Applied to the case of a commercial aircraft operating voice and VDL 4 on opposite side antennas with 53 db of isolation (50 db antenna coupling and 3 db cable loss) between interferer and victim receiver, with a frequency separation of 40 channels (1 MHz) between voice and VDL Mode 4, the following maximum protected distance from the desired data source are derived: Channel Load Criteria 10% 20% 30% 40% 50% 60% 70% 80% 95% within 1,50 sec (1a) ,996% within 3,75 sec (1b) % within 3,75 sec (2a) ,996% within 11,25 (2b) % within 7,5 sec (3a) 35 99,996% within 15 sec (3b) % within 22,5 sec (4a) 99,996% within 45 sec (4b) Table 5 - MPD descend flight phase - Commercial aircraft Opposite side antennas default VDL Mode 4 parameters Short procedure A cell filled with means that the criterion is fulfilled independently of the distance to the desired signal source. All other cells are presenting a protected distance smaller or equal to 35 NM, which puts an operational constraint. Only the Short procedure has been analysed since with the default parameters an important part of the performance criteria cannot be met Large aircraft, recommended set of VDL Mode 4 system parameters (PS4) This time the same aircraft and frequency separation is considered, but assuming the VDL Mode 4 system is operated making use of recommended Parameter Set PS4 (detailed in D3, Appendix C.5). With these parameters, it appears that the short procedure can meet most of the criteria. The long procedure included in the VDL Mode 4 standard has thus also been analysed. Edition: V4.1 Page 23

24 Channel Load Criteria 10% 20% 30% 40% 50% 60% 70% 80% 95% within 1,50 sec (1a) ,996% within 3,75 sec (1b) % within 3,75 sec (2a) 25 99,996% within 11,25 (2b) 25 95% within 7,5 sec (3a) 0K 99,996% within 15 sec (3b) N0K 95% within 22,5 sec (4a) 99,996% within 45 sec (4b) Table 6 - MPD descend flight phase - Commercial aircraft Opposite side antennas Results with recommended Parameter Set PS4 Short procedure Channel Load Criteria 10% 20% 30% 40% 50% 60% 70% 80% 95% within 1,50 sec (1a) N N N N N N N N 99,996% within 3,75 sec (1b) N N N N 95% within 3,75 sec (2a) N N N 99,996% within 11,25 (2b) N 95% within 7,5 sec (3a) N 99,996% within 15 sec (3b) % within 22,5 sec (4a) 99,996% within 45 sec (4b) Table 7 - MPD descend flight phase - Commercial aircraft Opposite side antennas Results with recommended Parameter Set PS4 Long procedure Category number In summary, considering the present case of large aircraft, the set of time-critical criteria are partially respected as follows. The maximum load considered is 70%. Message Category Initial results _Default parameters Criteria (a) 95% Criteria (b) % Short procedure (Parameter Set PS4) Criteria (a) 95% Criteria (b) % Long procedure (Parameter Set PS4) Criteria (a) 95% Criteria (b) % 1 Distress (V) Urgent (H) Flight Safety (M) Edition: V4.1 Page 24

25 Category number Message Category 4 Routine Surveillance or Navigation (L) Initial results _Default parameters Criteria (a) 95% Criteria (b) % Short procedure (Parameter Set PS4) Criteria (a) 95% Criteria (b) % Long procedure (Parameter Set PS4) Criteria (a) 95% Criteria (b) % Red=Not fulfilled -Yellow = partially fulfilled (Channel load < 60%) -Green = Fulfilled Black = Never fulfilled (even without interference The results show an important improvement using the recommended parameter set PS4 over the default VDL Mode 4 parameters. Within the time-critical defined levels: L level is respected on two criteria by short and long procedures; M and H levels on two criteria by short procedure only; V level (Distress) criteria are not met. Other aircraft size (other antenna isolation values) and frequency separation cases also should ideally be considered, up to the worst operational case (e.g. small aircraft with 35 db of isolation between antennas, and with the 2 nd or 5 th channel adjacent to voice to be used for VDL Mode 4) Applied to ADS-B applications Here the aircraft is assumed to receive ADS-B signals from relatively distant stations (transmitter in air or ground) and whose rather weak signal are interfered by aircraft own voice transmissions. Applications like TIS-B, and air-to-air ADS-B like ASAS, traffic sequencing and merging, are examples where airborne co-site impact is considered most likely critical. A usual requirement applied as Quality of service for ADS-B is the requirement of 95% successful transmissions. Since ICAO VDL Mode 4 criteria is a MER not greater than 2% (BER<1.10-4), this figure has also been used. A short analysis is made here for two cases of commercial aircraft, with the minimum isolation guaranteed between antennas by major airframes manufacturers (53 db for opposite sides antennas, 38 db for same-side antennas, including 3dB feeder loss) The tests were achieved at 119, 128 and 136 MHz. The 5 th, 20 th and 40 th adjacent channels have been tested for the interferer. For each scenario, the maximum protected distance between the desired signal transmitter and the receiver is given in NM on the basis of the above criteria (95% or 98% reception success There is no retransmissions here, unlike in point to point transmissions). Edition: V4.1 Page 25

26 The results are presented in the tables below: Criteria A MER < 5% Criteria A MER < 2% Interferer adj. channel # 5 th 20 t f 40 th 5 th 20 th 40 th # Interferer Flight Phase 1 OPC Departure OPC Climb 3 OPC En-route 4 OPC Descend OPC Arrival ATC Departure ATC Climb ATC En-route 9 ATC Descend ATC Arrival OPC+ATC Departure OPC+ATC Climb OPC+ATC En-route 14 OPC+ATC Descend OPC+ATC Arrival Table 8 - Max. Protected Distance - Commercial aircraft Opposite side antennas -ADS-B Criteria A MER < 5% Criteria A MER < 2% Interferer adj. channel # 5 th 20 t f 40 th 5 th 20 th 40 th # Interferer Flight Phase 1 OPC Departure < 5 NM < 5 NM < 5 NM 2 OPC Climb 3 OPC En-route 4 OPC Descend < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 5 OPC Arrival < 5 NM < 5 NM < 5 NM 6 ATC Departure < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 7 ATC Climb < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 8 ATC En-route 9 ATC Descend < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 10 ATC Arrival < 5 NM < 5 NM 11 OPC+ATC Departure < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 12 OPC+ATC Climb < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 13 OPC+ATC En-route 14 OPC+ATC Descend < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM 15 OPC+ATC Arrival < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM < 5 NM Table 9 - Max. Protected Distance - Commercial aircraft Same- side antennas - ADS-B In order to figure out what the presented MPDs means from an operational point of view, they have been translated in the below table into delay, considering 2 aircraft flying the same speed. (400 km/h and 1000km/h) Edition: V4.1 Page 26

27 Distance (NM) Distance in time (2*400km/h) Distance in time (2*1000 km/h) 1 8s 3s 2 17s 7s 5 42s 17s 10 1min 23s 33s 15 2min 05s 50s 20 2min 46s 1min 07s 25 3min28 1min 23s Table 10 - Distance to delay translation for 2 aircraft converging at the same speed It appears from Table 8 that, with opposite side antennas, one can satisfy a 95%- reception criteria within 25 NM from signal sources, and from one MHz of frequency separation onwards. Such 25 NM is equivalent to a few minutes of flight for two aircraft approaching each other as shown in Table 10. For same-side antennas, with one MHz of separation, the 95%-reception cannot be fulfilled across a whole flight, except for signal sources at very short distance from the aircraft. Such small distances are equivalent to tenths of seconds of flight for two aircraft approaching each other. In conclusion, and based on the observed interference rejection presented by the available avionics, the airborne co-site interference from voice is presenting a strong constraint for ADS-B applications involving data-in streams to aircraft with the exception of the en-route case which is fulfilled in all conditions Applying measured rejection performance in simulations In the present section, airborne co-site interference is inserted in traffic simulator, and applied to a single aircraft. The goal is to search whether the co-site interference is visibly impacting the VDL quality of service, for point-to- point applications, and whether it already sets a limit that can be identified for a single aircraft on a VDL channel 5. The analysis is thus achieved for a single aircraft running point-to point data exchange, but only the air-ground round-trip transmission is simulated, without the ground to-ground segment to ATC or airline AOC- terminals. The set of CPDLC messages used consist of a medium-to-heavy use of CPDLC for which usually a requirement of 95%-transmission success within a certain time is required (eg 10 sec for round-trip in the case of VDL Mode 2). Analysis is made also for more demanding performance levels like those for Time-Critical applications. The simulated aircraft is positioned at high altitude, at 50 NM from the transmitting VDL ground-station. For the airborne co-site phenomenon, the VHF voice and VDL antennas are isolated by 50 db (+ 3 db of cable and connectors) and a frequency separation of 1 MHz is assumed. The interfering voice traffic is made of typical ATC and OPC voice traffic, and traffic volume is set as observed in ENR or DESCEND phases of flights, depending of the lines. The DESCEND phase is the one whose 5 A similar analysis had already been applied in co-site assessment for VDL Mode 2, ref[14] Edition: V4.1 Page 27

28 statistics show the highest voice traffic, this is reflected in the data-link performance. In the last two columns, a comparison is also made between a case with normal CPDLC and voice traffic and one extreme case with CPDLC traffic multiplied by 20 and interfering voice traffic multiplied by a factor M. The goal is to show that the limit can indeed be identified by simulations. The analysis shows that: For both VDL Modes, the performances of point-to-point applications are affected by airborne co-site interference. Not taking that into account results in over-estimating VDL performance. For default VDL 2 parameters, the performance limit per aircraft has no practical impact for CPDLC like applications: a 95% round-trip requirement within 7 sec is possible for an aircraft affected by co-site interference from voice. CPDLC is only operated in ENR phase but even in DESCEND the limit is not reached. Only climbing to a voice traffic multiplied by 30 (ENR) and 5 (DESCEND) would reduce the quality of service below this criteria, which is not a practical case. As observed by studies on VDL 4 capacity, the default VDL 4 parameters are resulting in low performance compatible with a part of time-critical applications criteria. The co-site phenomenon further reduces the VDL 4 performance but it is not the main VDL4 performance issue. With modified parameters (PS2 analysed in 2006) VDL 4 -performances are visibly improved 6. But a performance requirement like 95% Round-Trip within 2 sec is not possible for a single aircraft when affected by airborne cosite interference in DESCEND phase of flight). Illustrating again the co-site impact, one can observe that a single aircraft without co-site can fulfil on VDL 4 all presented criteria. When co-site is accounted, 95% -round-trip within 7 sec is still possible for a single aircraft in ENR or DESCEND; a limit can be detected by simulations, and VDL 4 is clearly beyond this limit in last column with CPDLC traffic multiplied by 20, and a voice traffic multiplied by 10 (in ENR), or not multiplied (M=1, DESCEND). 6 Although present document has been updated with PS4 Parameter Set, PS4 performs less efficiently than PS2, when considering co-site interferences. PS2 is thus preferred when drawing conclusions. Edition: V4.1 Page 28

29 Round-Trip delay VDL 2 No Cosite VDL 2 En Route - Cosite VDL2 Decend - Cosite Impact of airborne co-site interference on VDL quality of service for point to-point applications. (default VDL2-VDL 4 ICAO parameters except PS2-lines) 95% < 10s 95 < 5s 95% < 2s 98% < 20s % < 15s % < 5s 95% < 7s 95% < 7s CPDLC x 20 Voice x M 100% 100% 100% 100% 100% 98.2% 100% 98.8% NOT FULFILLED 89.2% 100% 100% 100% 100% 100% 100% 100% 100% NOT FULFILLED 98.2% 100% 100% 99.7% 100% NOT FULFILLED M= % NOT FULFILLED M=5 93.6% VDL4 No Cosite Default param. 95.5% NOT FULFILLED 41.5% NOT FULFILLED 16% 100% 100% NOT FULFILLED 37.4% NOT FULFILLED 64% NOT FULFILLED 0% VDL 4 En Route Cosite Default parameters VDL 4 No Cosite - PS2 parameters VDL 4 En Route Cosite PS2 parameters VDL 4 Descend Cosite PS2 parameters NOT FULFILLED 92% 100% 100% 99.7% NOT FULFILLED 38% 100% 98.9% 96.7% NOT FULFILLED 10% 100% 98% NOT FULFILLED 63% 100% 100% 100% 99.94% 100% 100% 100% NOT FULFILLED 99.91% NOT FULFILLED 35% 100% NOT FULFILLED 98.5% NOT FULFILLED 96.6% NOT FULFILLED 50% 100% 100% 98.8% NOT FULFILLED M=1 0% 98% NOT FULFILLED M= % NOT FULFILLED M=1 44.3% Table 11 - Impact of airborne co-site interference on VDL data-link quality of services Edition: V1.0 Page 29

30 4. CONCLUSIONS OF PRESENT DELIVERABLE Based on laboratory tests applied on one ground-station and one avionics, the present deliverable has produced some interference testing results resulting in a proposal for separation requirements for VDL 4 operation in the VHF aeronautical band. This time the previous issues observed on the avionics radio have been solved and the proposed criteria would be feasible for implementation. The study also produced audio samples of worst and average cases of airborne cosite VDL 4 interference on voice reception in adjacent VHF channels. The present study also analysed the rejection performance of VDL 4 receiver against interfering co-site voice traffic based on the measurement of one available avionics The measured rejection performance has then been used in an analytical way to assess which point-to-point applications (from CPDLC on, up to most-demanding time-critical performances levels) were feasible without setting strong operational constraints in frequency separation or maximal distance to ground-stations. It appears that taking the airborne co-site interference into account, VDL Mode 4 is compatible with a part of time-critical point-to-point applications criteria, although the main performance limitation is not the one resulting from the airborne co-site interference. With a similar approach for ADS-B, it appears that the airborne co-site interference from voice is presenting a constraint in range and frequency separation for ADS-B applications involving data-in streams to aircraft. The maximum protected reception distance has been evaluated in the case of commercial aircraft with two levels of antenna isolations and in different flight phases and interfering voice traffic conditions. The application of the new parameters set (analysis made in 2009) is not subject of change in ADS-B application. For Point -to -point system performances, the new parameters set importantly improves the resistance to cosite interference. A larger part of the defined timecritical point-to-point applications criteria is compatible (eg up to 95% success within 3.75 sec in most channel load conditions) in the case of VDL Mode 4 short procedure messages. Long procedure transmissions are not compatible with the most demanding criteria. Edition: V1.0 Page 30