38TH ANNUAL CONFERENCE, Santiago, Chile, 15-19 March 1999
WP No. 102
The Definition of “Direct” in Direct Controller Pilot Communication (DCPC)
IFATCA has policy on the use of datalink communications, and includes the following reference to “Direct”:
|“In any ATS system, where data link is considered a safety-critical element of that system, data link based ATS must be accompanied by direct two-way controller-pilot voice communications which is also safety critical. This direct voice functionality shall be rapid, continuous and static free.”
“The definition of direct as used in voice or data link communications requires that no third human party is involved in the set up and/or delivery of these communications. Any set up procedures by either the pilot or the controller must be minimal and nearly instantaneous.”
This definition effectively covers the case where a third party service provider is used. However, there remains ambiguity when IFATCA representatives are trying to obtain endorsement of this IFATCA policy in the various working groups and panels that are working on applications of digital communications.
This paper addresses the ambiguity directly, and thus extends the IFATCA position and policy further.
There are a number of technologies available in the Navigation and Surveillance domain’s that are able to support a reduction in separation standards. In working towards these reductions in separation standards the communications domain has become significant. Communications are identified as one of, if not the major, constraint in the plans to reduce separation standards.
In operational environments where HF radio is used, any move to digital air-ground communications, such as datalink have clear advantages in usability. In terms of performance, there are distinct advantages provided that the performance of the digital air-ground communications system meets any defined performance criteria.
ICAO documentation provides some insight into the requirements for reducing non-radar separation in such operational environments, and communications performance:
(a) When special electronic or other aids enable the pilot-in-command of an aircraft to determine accurately the aircraft’s position and when adequate communication facilities exist for that position to be transmitted without delay to the appropriate air traffic control unit; or
(b) When, in association with rapid and reliable communication facilities, radar-derived information of an aircraft’s position is available to the appropriate air traffic control unit.
From this it can be deduced that the descriptors of communications, the performance targets that must be met are: “adequate”, “rapid” and “reliable”.
Experience has shown, and provided to SC1 by IFATCA members, that the present generation of datalink communications are not rapid and reliable.
A number of problems have been experienced by pilots as well.
When considering VHF digital air-ground communications, the problem becomes more complex. VHF air-ground communications are currently carried out by VHF or UHF radio telephony. The characteristics of VHF/UHF RT are quite different from HF – it is a means of communication that is less susceptible to the problems that affect human perception of voice communications such as noise and distortion of the signal.
The ICAO ATS Planning Manual identifies communications as “a specific circumstance which may require the application of greater radar separation than the minimum normally prescribed” noting that communications congestion is a radar controller limitation. PANS-RAC refers to communications obliquely, but considers direct speech facilities, in the case of the transfer of radar control, as being available at all times, which permit communications to be established instantaneously. In this case,“Instantaneous refers to communications which effectively provide for immediate access between controllers”.
In addressing the issue of Direct, it is the latter that provides the key.
What is immediate access? What factors have an affect upon datalink communications so as make its performance unacceptable in terms of “immediate access”?
There are three levels which can have an affect this attribute:
(a) Transaction time
Transaction time: The transaction time is the time it takes for a datalink communication to pass from message composition through transmission to receipt by the recipient, and all associated parity checks, i.e., when there is acknowledgement that a message has been received.
Interface: The interface refers to the design of the human computer interaction that supports, or enables a dialogue between the pilot or controller and the datalink system.
Interaction: This refers to the behaviour of the human-machine system, that is the way that the datalink system influences the behaviour of the controller and pilot when actively using the datalink, and must be seen in the context of the overall task that the pilot or controller are carrying out.
The present datalink communication systems fielded must be considered as “first” generation. It is difficult to make conclusions from these, however, it is possible to make generalisations about what we know of datalink communications in the context of the three factors above from the systems in operation, and also form simulation studies.
Transaction times in operation have on occasions led to delay in messages being received. Data collected from Tokyo ACC shows:
1) 1500 CPDLC messages were exchanged during the trial period.
2) The average delivery time for CPDLC downlink was 29 sec, Maximum delivery time was 6 min 52 sec, and standard deviation was 32 sec.
3) The average delivery time for CPDLC uplink was 63 sec, maximum delivery time was 4 min 55 sec, and standard deviation was 56 sec.
In general, visual datalink reduces the transaction times, particularly where a message may contain long or multi-point clearances. This is a benefit over voice communications. However, it is also clear that visual workload increases, and that the ability to multi-process information cognitive reduces because of the reliance on a single sensory channel. This is because of a behavioural, or interaction, response to datalink communications. The timing of datalink messages can contribute to a disruption in the controllers workload, particularly when there are small intervals between messages, and the unread message count rises – this can lead to the human being “driven” by the system, with, consequentially, a risk that a controller will not fully attend to the messages, merely seek to acknowledge them. If a message is sent, and a response is awaited, this may cause a diversion – of effort. In some cases slow transaction times can lad to serial rather than parallel information processing by controllers – which is one of the major human strengths in the control process as we know it.
The interface has caused problems. The more difficult an interface, then the greater the mental workload. The menu driven datalink interfaces assist in mitigating the workload, but where non- standard messages are involved, then this can add to workload, the RTCA DO219 message set has found to be lacking in operational experience.
Consequentially – where mixed voice and datalink mediums exist, it has been shown that crews respond to each sequentially – queuing both voice and datalink into one single queue. In tactical environments such as that where radar is employed, separation is only deemed to exist when a pilot has acknowledged an instruction. If a delay occurs in acknowledging a tactical instruction, then the delay can ripple across other tasks that a controller has to carry out, slowing down the rate at which they can be carried out. Indeed it is the effect that datalink communications has upon task scheduling and prioritisation that causes the most concern. That these concerns exist at all, is indicative that datalink communications, at the present stage of development are not “direct” in the context of “adequate”, “rapid” and “reliable”, and to communications which effectively provide for immediate access.
Communication, by its very definition means a two way dialogue – i.e. a stimulus and a response. if no response is given, then the communication dialogue remains open. In today’s voice environment, this response is direct, is immediate, if no one responds by return, then the controller can query or challenge the pilot. Equally, the pilot can make a request, and if nor response is obtained, the pilot can challenge the controller, by return. Datalink communications may offer this, but it the form that the response may take can be diverted to one of several visual display mediums – the track data block on a radar display, the updating of a label on an ADS display, the highlighting of a field in an aircraft’s active flight plan on a tabular list. This involves visual searching in these displays for the appropriate item to indicate the response. Voice communications, because it is a different sensory modality from the visual one, allows a controller to attend to visual cues, as well as to attend to aural ones. Datalink using visual only, means a potential reduction in the cognitive capacity of the controller, and thereby affects the immediacy of the communications as perceived by the pilot or by the controller.
When considering the performance of a datalink system as being “direct” controller pilot communications, there are three factors that influence its ability to be direct. Because of the effect of these factors upon human performance and behaviour, failure to implement systems without these factors having been solved in the design, will lead to the system as being unable to support procedures such as reduced separations.
For digital air-ground datalink communications to be considered to be direct controller pilot communications, they shall have demonstrated that they support communications dialogues that provide equivalent VHF voice characteristics in terms of the transactions times, the interface which the human uses to interact with the system, and in terms of the cognitive interface between human and system.
IFATCA Technical manual Sec.6 Chap 2-1 Direct Controller Pilot Communication.
ICAO ATS Planning Manual, Para 2.8.
ICAO PANS-RAC, Para 2.8.
Statistical Data from Japan CPDLC trial.
Last Update: September 28, 2020