From Analogue Broadcast Radio Towards End-to-End Communication

From Analogue Broadcast Radio Towards End-to-End Communication Horst Hering *, Konrad Hofbauer *+ * EUROCONTROL Experimental Centre, Brétigny, France + Graz University of Technology, Austria The capacity of the current ATC system is, among other factors, limited by the maximum number of aircraft that a controller can handle in a sector. This led in the past to a decrease of sector sizes in order to increase capacity. We study in this paper the impact of small sectors on the air-ground radio communication. Small sectors require a large number of radio channels, and the sector handovers generate multiple radio calls, which are workload for both controllers and pilots. We outline in this context an initial idea to make the control sectors transparent for the aircrew. With a network of radio base stations and reduced transmission powers a cell-based end-to-end communication system can potentially be established, without changing from analogue to digital radio. The aircraft transmits on the same frequency across all sectors, and the ground network routes the voice calls to the appropriate controllers. We briefly discuss potential benefits and issues of this concept and see a clear need for further research to determine the feasibility of this idea. I. Introduction The convention on international civil aviation (also known as Chicago convention [1, Annex 10]) was signed in 1944, and is the base for current standards and recommended practices in air-ground communication. The standardised amplitude-modulation (AM) radio is in operation worldwide and has basically remained unchanged. A. Voice Radio Communication Technique in Use The AM radio is based on the double-sideband amplitude modulation (DSB-AM) of a sinusoidal carrier. For the continental air-ground communication, the carrier frequency is within a range from 118 MHz to 137 MHz, the very high frequency (VHF) band, with a channel spacing of 25 khz (~700 VHF voice channels). Due to VHF channel shortage in Europe some European areas are operated with a reduced channel spacing of 8.33 khz. The standardised technology has a very small speech frequency bandwidth (300 Hz to 3.4 khz) and is known for its poor transmission quality. For the 8.33 khz channel spacing the speech frequency bandwidth was further reduced to 350 Hz to 2.5 khz. Poor transmission quality and the narrow speech-bandwidth decrease the intelligibility of the transmitted speech. The aeronautical radio transmits with an AM carrier power in the range of 30-50 W. The transmission power allows save radio communication over a few hundreds of nautical miles, given that in most situations there is a direct line of sight between transceiver ground station and aircraft. The reuse of the same channel frequency requires a large geographical separation (~1000nm), which limits the set of radio channels that can be used within a given area. The number of available VHF voice communication channels is restricted. To avoid an expected shortage, a 8.33kHz channel spacing was implemented in some European areas in 1999. Despite this effort, EUROCONTROL predicts a further shortage of VHF frequencies in Europe for 2015 and onwards [2]. B. Operational Concept In the early days of commercial air transportation, electronic equipment for communication, navigation and surveillance was expensive, power consuming, large and heavy. The standard for airborne equipment was limited to light radios for communication and navigation, while more heavier equipment for navigational aids and surveillance were ground-based. Therefore the concept of blind aircraft pilots guided by a seeing ground controller arose. The VHF radio was and is the main tool for the controller to give flight instructions and clearances to the pilots. This broadcast radio communication is usually called a "party-line", used on a time-shared base by one 1

air traffic controller and all aircraft in the corresponding flight sector. It is the controller s responsibility to safely separate the aircraft trajectories, following defined rules. C. What will be the Future of Air-Ground Communication? Since the shortcomings of the current communication concept have a significantly different impact in North America and Europe, there is presently no international high-level agreement for a long-term solution. However, the ongoing international strategy discussions show a tendency towards a reduction of conventional air-ground voice communication, and instead favour digital data communication using a data link technology. Figure 1 shows the common FAA EUROCONTROL Action Plan for the future air-ground communication strategy. Therein, VHF voice communication (incorporating digital voice) persists beyond 2025, and planning for a future VHF link technology is not started before 2022. The European Commission and EUROCONTROL launched in 2007 the Single European Sky ATM Research program Figure 1. Future Air-Ground Communication (Source: FAA/EUROCONTROL AP17) (SESAR). Its operational concept is mainly based on digital data communication, but situations will remain in which clearances and instructions are issued by voice [3]. II. Issues with the Current Broadcast Communication System Beside the previously stated technical issues such as the poor channel quality and the narrow speech-bandwidth, there also exist important operational constraints. We address in the following mainly those issues that are related to the small sector sizes in certain areas of Europe. Many of these sectors are crossed by a modern aircraft within five to eight minutes. Small sectors, and the resulting frequent sector changes, interfere with the currently used communication concept. A. Shortage of Radio Channels Each sector requires an associated sector frequency (channel). As described in Section I, the reuse of the same frequency requires a very large geographical separation, and a channel shortage for Europe by 2015 is predicted. B. Sector Change Related Speech Acts The VOCALISE study [4] of the French DSNA (Direction des Services de la Navigation Aérienne) examined 60 hours of air-ground voice communication in twelve distinct French enroute sectors featuring heavy traffic periods. The results show a mean of 324 speech acts per hour, with an average channel occupation of 24 minutes/hour (40%). The channel occupation rose up to 90% during 5 min. peak periods. VOCALISE reports that about 60% of the speech acts are fully or partially related to sector change procedures. A EUROCONTROL Experimental Centre note showed very similar results with the analysis of 9138 speech acts of Swiss and German en-route controllers [5]. Figure 2 shows the overall distribution of the different types of controller instructions. The second horizontal axis gives the number of instructions in a single speech act. The transfer group consists of controller speech acts for aircraft leaving and coming into the sector. Figure 2. Distribution of ATCO Instructions (Source: [5]) 2

C. Sector Changes - a Risk Factor for Loss of Air-ground Communication Sector changes are a potential cause for an interruption of the radio communication with the assuming control sector. For the transfer of an aircraft to the consecutive sector, a controller transmits a voice message on the sector frequency, for example: Lufthansa tree four niner contact Bremen radar on frequency one two six decimal six five. Several factors, such as human s imperfection in speaking and understanding, low VHF transmission quality, pilot s delay in the execution of the frequency change, or human errors might lead to the frequency not being changed or the wrong frequency being selected. In such a case the aircraft enters a sector without radio contact to the responsible controller. This incident creates at least supplementary workload for the pilot and the sector controller, but may also cause a hazardous situation. A EUROCONTROL study [6] reports as highest contribution factors in loss of air-ground communication occurrences radio interference with 29%, change of the frequency with 25%, and all other reported reasons with less than 3%. D. Other Related Issues 1. Call-Sign Problems Call-sign mishearing, misunderstanding and call-sign confusion are an important issue in ATC safety. A recent EUROCONTROL study [6] showed: Incidents involving air-ground communication problems between controller and pilots are rare and encompass about 1% of all reported incidents and 23% of ATC related incidents. 2. Task Load Generated by Party Line While almost all broadcasted messages heard at the ground station are addressed to the controller, the pilots have to identify in the stream of message those calls that are addressed to their aircraft. This is a non-trivial mental task for the pilots, especially since the aircraft address (call-sign) that they have to respond to changes with every flight. It is thus common cockpit practice that the pilot non-flying fulfils the communication task. 3. Task Load Generated by Sector Changes Operational experience with the current air traffic management (ATM) concept shows that the human operator represents an important capacity bottleneck for the overall ATM system. The number of aircraft a controller can handle safely at the same time is limited by human s mental capacity. In the past, in order to increase the overall ATM capacity, the sector volumes have been reduced such that the number of aircraft in a sector is held within limits acceptable for the human operator. The frequent sector changes now generate a significant task load for controllers and pilots. 4. Controller Pilot Data Link Communication Controller Pilot Data Link Communication (CPDLC) is explored as remedy for the communication issues. CPDLC complements the analogue voice channel with a digital data channel. Up-linked transfer messages will be less error-prone than voice messages, but the underlying tasks for the sector change are persisting for pilots and controllers. III. Advanced End-to-End Communication for ATC Air-ground voice communication can be seen as one of the bottlenecks for a future increase of the current ATM capacity. Small sectors increase the number of transfer and assume messages, which generate about 60% of the communication volume and are prone to human errors. Making sectors transparent for the airspace users would reduce the number of communication acts and eliminate the sector-change related sources for human errors. A. Basic Concept We propose to investigate an end-to-end communication concept for ATC air-ground communication. With end-to-end communication we mean that a user s handset is transparently connected to another user s handset via a communication network. While the handsets may permanently change their positions, the network nevertheless guarantees an uninterrupted end-to-end connection. For the user the lower layers of the network should be invisible and transparent. The same principle is applied in public mobile telephony systems such as GSM (Global System for Mobile communications). The aircrew needs a permanent end-to-end voice communication channel to the responsible controller. This is valid everywhere and for any phase of the flight. In the current ATC concept a geographical airspace volume limits the responsibility of the controller. The airspace an aircraft is present in defines to which controller the end-to-end communication has to be linked to by the network. If an aircraft reaches the geographical limit of controller s responsibility, the controller hands over the symbolic handset of this end-to-end communication to the neighbouring controller. This means that the controller transferring an aircraft instructs the network to reroute the 3

end-to-end communication to the neighbouring controller. The aircrew does not need to notice this rerouting, which means that the ATC sectors become transparent for the aircrew. Hypothetically, end-to-end communication could be an answer to the existing bottleneck of in ATC air-ground communication, and a capacity increase for the overall ATM system could be expected. End-to-end communication may also be of interest in light of the predicted increase of business jet traffic with mostly single-pilot cockpits. B. SESAR and End-to-end Communication Also the European SESAR research program states that digital data communications may eventually obviate the need for discrete sector frequencies and associated frequency changes onboard, since communications will be addressed to an aircraft or ground station with the delivery method being transparent. The SESAR deliverable D3 from July 2007 [3] makes the following statement in section C7.1 : When traffic density increases beyond a certain level, the number of voice messages to be exchanged reaches a point beyond which it is no longer possible to ensure the timely passage of information between pilots and air traffic controllers. The SESAR concept of operations utilises digital data communication applications and services as the main means of communication. Digital data communication applications are not affected by voice frequency congestion. Digital data communications may eventually obviate the need for discrete sector frequencies and associated frequency changes on board, since communications will be addressed to an aircraft or ground station with the delivery method being transparent, however the workload implications of such a development and the loss of the benefits of a broadcast communication channel will require careful study. Addressing changes associated with the transfer of communications will be handled automatically. We conclude from SESAR s description of the communication principles that the future communication seen by SESAR is: Digital data communication Addressed communication Automatically transferred communication IV. Proposed Intermediate Technical Solution - Adapting the Current Broadcast Radio Communication In the following sections we propose the development of an intermediate technical solution and point out research and development directions on how to realise a quasi-end-to-end communication using the current voice communication standard. A. Speech Watermarking Techniques Advanced Digital Features for the Analogue Radio Communication By using so-called speech watermarking techniques, it is possible to embed digital information, such as aircraft call-sign, unique aircraft identification address or tail number, into the analogue voice communication. It is thus possible to transmit in conjunction with the voice transmission the identification of the aircraft. The transmitting aircraft can then be unambiguously identified by the ATC ground system by extracting the embedded aircraft identification. An embedded watermark is unnoticeable for humans in the received speech communication. In 2003 the embedding of a digital signature as watermark in the pilot s voice was proposed as Aircraft Identification Tag (AIT) [8, 9]. Figure 3 shows the basic principle of AIT. An AIT Initial Feasibility Study [10] was launched by the EUROCONTROL headquarter in 2006. The study reported no potential technical constraints for a realisation [11]. A study on new AIT embedding algorithms [12] reports data embedding rates up to 2000 bit/sec, which is higher than the rate of 100 to 150 bit/sec mentioned as required in the AIT - Initial Feasibility Study [10]. This feasibility study Figure 3. AIT Aircraft Identification Tag has foreseen to embed in the watermark among other data the originator and destination address of the voice message. 4

B. Cellular-Like End-to-end Communication Using the Current Broadcast Technology 1. Public Mobile Telephony Network Most of the current mobile phone networks are based on the GSM standard. One of the basic principles of this standard is a large number of local radio base stations that act as interface to the network of fixed and mobile phones. Thus the mobile handset requires a transmission power of only a few hundred milliwatts for the connection to the nearest base station. The geographic area that is covered by one base station is called a cell. A moving handset will automatically switch over to another cell if the edge of the cell is reached. The cell structure of the network is transparent for the user. 2. New Ground Structure for the ATC Radio Communication Similar to the mobile phone network, a large number of ATC transceivers could be deployed at regular distances. Each transceiver represents a cell of the hypothetical ATC ground network. The whole network operates at a single frequency, and all cells are connected to a management unit. Figure 4 shows a possible cell layout for such an ATC ground network. The distance between the cell and the aircraft equals approximately the flight level of the aircraft. The transmission power of the aircraft and the ground transmitters is reduced to a value that minimises interference between adjacent cells but still allows the safe reception of the signal. Cell receivers not directly below the aircraft are further away from the aircraft and will receive the transmitted aircraft signal with a lower signal level. Based on these level differences and using triangulation and/or bearing, the management unit can determine in which cell the transmitting aircraft is located. The following example shows the signal reduction over distance. Let us assume that the aircraft s transmitter antenna is an isotropic radiator, which radiates with the same power in all directions. The power density s o received at a point of the sphere is then directly dependent on the transmitted power P tx and the surface size A sphere of the sphere. s o = P tx A sphere = P tx 4"r 2 Simplified, for constant P tx and a constant surface of reception the power density s o will be reduced by the square of the distance r. Let us assume for a concrete example that an aircraft flies in the upper airspace at flight level 300 (30,000 feet = 5nm) and that the distances between the transceivers of the cells are 15 nm. The power received at the transceiver directly below the aircraft (Figure 5) is normalised to be one. The transceiver cells beside receive about 1 of the 10 normalised power (-10 db) and the next transceiver cells 1 beside receive of the normalised power (-15 db), 37 only. Figure 4. New cell structure for the ATC ground network Figure 5. Communicating with a single cell transceiver V. Originator and Destination Addresses for End-to-End Communications ATC air-ground voice communication follows established procedures. If possible, any modification of these procedures should be avoided. The addition of an originator and destination address should be automated as far as possible by the user s equipment or the network. 5

A. End-to-End Communication Transmitted by the Aircrew For the aircrew, each transmitted end-to-end communication is addressed to the responsible controller. The addresses embedded in the voice message are: Originator address: aircraft identifier Destination address: ATC controller. The network cell receiving the aircraft call is connected to the responsible controller for this network cell. Figure 6 shows an example for a possible sectorisation. In the example, the aircraft is currently controlled by the green sector. When the aircraft has reached the blue cell, the ground network automatically Figure 6. Sectorisation with a cell structure reroutes the aircraft voice communication to the controller of the blue sector. This rerouting could potentially be transparent for the aircrew. B. End-to-End Communication Transmitted by the Controller At the controller s side addressing is less straightforward, as a controller holds many ends of end-to-end communications. The originator address is simply the ATC controller, but the destination address needs to be provided to the network. With the destination address the network is then able to identify the network cell that is straight below the addressed aircraft. The current controller working procedure for calling an aircraft would have to be changed, as the controller has to indicate the destination address of the aircraft. For example, the controller could point with a mouse pointer to the label of the addressed aircraft on the radar screen, press the PTT (Push To Talk) switch, and start to speak. From this information the display system could extract the destination address and the current position of the aircraft, and therewith the network cell can be addressed. To collect controllers opinion on such a change of the calling procedures the EUROCONTROL Experimental Centre made an initial usability study. The study showed that such a change could be acceptable for the controllers [13]. VI. Benefits and Issues of the End-to-End Communication A. Issues of the End-to-End Communication Following concept issues are identified and require further study: Controller agreements for geographically flexible aircraft handovers are not supported. The sector-wide party-line is reduced to a local-area party-line. As no sector-wide party-line exists, simultaneous calls of multiple aircraft must be queued. B. Benefits of the End-to-End Communication The proposed concept interconnects aircraft transceivers and the nearest cell transceiver of a new ground network using low power radio transmissions. This allows the current party-line based communication concept to move towards an end-to-end communication concept similar to the cell based mobile telephony system. The benefits of an end-to-end like radio communication for ATC are: Makes ATC structure and sectors transparent to pilots Avoids shortage of communication channels Eliminates frequency change task for aircrew Prevents loss of communication related to frequency change Reduces controller workload related to sector change voice messages 6 Figure 7: ATC transmits watermarked voice with embedded destination address (source [13])

Eliminates aircrew s party-line monitoring task Beside the above benefits, the end-to-end radio communication could be an enabler for new operational concepts. For example, the size of a sector becomes an ATC internal issue, as it is invisible for the aircrew. Consequently a controller could be responsible for a group of aircraft on a future highway, while another controller controls another group of aircraft on the same highway. The concept also includes the AIT concept and therewith all its benefits for safety and security. VII. Conclusion It is internationally agreed that a change towards digital radio communication is required. The digital communication would most likely implement an end-to-end radio communication concept. Due to different constraints, a digital standard and a time scale for its implementation are not yet defined. This paper presented innovative concept ideas, which potentially allow the current analogue broadcast voice communication to move towards an end-to-end oriented communication concept. This would prevent frequency shortage and support novel operational concepts with digital communication features at an earlier time than it is foreseen for digital communication to be implemented. Further on, end-to-end communication would make the ATC structure transparent for the aircrew. The EEC launched human factor studies to identify the workload for aircrews and controllers related to sector changes. These studies may quantify how far the expected workload reductions influence the overall control capacity, and results are expected for the end of 2008. The proposed concept, in combination with AIT, brings digital features to the legacy analogue radio communication and allows an operation similar to a digital end-to-end manner. It represents an intermediate step towards the radio of tomorrow using technical standards of yesterday. VIII. References [1] http://www.icao.int/icaonet/dcs/7300.html [2] ICAO, Future Aeronautical Mobile Communications Scenario, Appendix A to Report on Agenda Item 2, 8th Meeting of AMCP, Montreal, February 2003. [3] SESAR Concept of Operations D3, July 2007, http://www.eurocontrol.int/sesar/public/subsite_homepage/homepage.html [4] L. Graglia et al., VOCALISE: Assessing the Impact of Data Link Technology on the Radio Telephony Channel, in proceedings of the 24th Digital Avionics Systems Conference (DASC 2005), Washington DC, USA, 2005 [5] H. Hering, Technical analysis of ATC controller to pilot voice communication with regard to automatic speech recognition systems, EUROCONTROL Experimental Centre Note 01/ 2001, Bretigny, France, 2001 [6] G. Van Es, Air-Ground communication safety study: Analysis of pilot-controller occurrences, Ed.1.0, EUROCONTROL, DAP/SAF, 16.04.2004 [7] CASCADE Stream 1 Real-time simulation, EUROCONTROL Experimental Centre Report 404, 2006 [8] H. Hering, M. Hagmueller, and G. Kubin, Safety and security increase for air traffic management through unnoticeable watermark aircraft identification tag transmitted with the VHF voice communication, in proceedings of the 22 nd Digital Avionics Systems Conference (DASC 2003), Indianapolis, USA, 2003. [9] M. Hagmueller and G. Kubin, Speech watermarking for air traffic control, EUROCONTROL Experimental Centre, EEC Note 05/05, 2005. [10] AIT- Initial Feasibility Study, EATMP Info-centre, EUROCONTROL Brussels, Belgium (2006) [11] M. Sajatovic et al., AIT study equipment and implementation scenarios, in proceedings of 7 th Integrated Communications, Navigations, and Surveillance, Washington DC, USA, 2007 [12] K. Hofbauer and H. Hering, Noise robust speech watermarking with bit synchronisation for the aeronautical radio, in Information Hiding, vol. 4567/2008 of Lecture Notes in Computer Science, pp. 252 266, Springer-Verlag, 2007. [13] H. Hering, K. Hofbauer, Towards selective addressing of aircraft with voice radio watermarks, in proceedings of the 7 th AIAA Aviation Technology, Integration, and Operation (ATIO) conference, Belfast, UK, 2007. 7