The data link research conducted by the ATOPS program was an effort to improve another aspect of ATC system operations. In addition to causing traffic delays, the growing congestion in terminal airspace was creating a radio frequency overload problem. So many pilots were trying to talk to controllers on the same frequency that it was often difficult for a pilot to squeeze in a transmission or request. In addition, the rapid pace at which controllers had to read off clearances and instructions in busy situations often led to miscommunications, errors, and the need to repeat transmissions. One possible solution the ATOPS program researchers saw for these problems was a twoway data link system.

A data link system would allow messages between pilots and controllers to be displayed on CRT screens in the cockpit and at the controller's station. The messages could be read and reread as necessary and stored for future reference. As a result, the system might be able to reduce not only the congestion of ATC frequencies, but also the number of miscommunications and errors and the need for repeated transmissions. [Ref 7-17] With the use of a satellite network, a data link could also allow pilots to communicate with controllers from remote locations, such as over the Pacific Ocean. In addition, data link offered the possibility of enhanced capabilities, such as displaying realtime weather reports and charts in the cockpit and direct communication between the airplane's flight computer and computers on the ground.

Engineers at the Langley Research Center began investigating the use of a twoway data link system for ATC communications in the early 1980s. Researchers working on a Single Pilot IFR (SPIFR) project conducted a flight test in a light twin engine airplane to evaluate whether a data link system would make ATC communications easier for pilots of small airplanes flying under instrument flight rules (IFR). During the SPIFR flight test, a pseudocontroller in the back of the airplane composed ATC messages for the pilot, which were then sent to the front of the plane and displayed on a screen in the cockpit. [Ref 7-18]

The flight test was followed by a series of realtime, piloted simulation studies that examined the use of various levels of data link capability in single pilot IFR flight operations. The simulation research showed that the data link system and cockpit display designed by the Langley engineers did, in fact, lessen the demands on the pilots' short term memory and reduce the number of transmissions needed between the pilots and air traffic controllers. The results also indicated that a data link system made it easier for the pilot to allocate time to critical cockpit tasks while receiving ATC messages. [Ref 7-19] Encouraged by the positive response they got from the single pilot flight and simulation studies, researchers began to look at data link technology for ATC communications in jet transports.

A commercial VHF data link system, called the ARINC Communications Addressing and Reporting System (ACARS) was already in use by the airlines. It allowed airline dispatchers to relay company messages, weather, and flight plan information to pilots. By 1989, Canadian air traffic controllers were also using an ACARS data link to transmit oceanic clearances to cockpit printers in aircraft preparing to cross the Atlantic Ocean, although pilots had to read back the clearances aloud over a conventional radio frequency. In 1991, the FAA began to provide predeparture clearances via an ACARS data link to airliners while they were still parked at airport gates. [Ref 7-20] Using a twoway data link as the primary communication mode for tactical and strategic ATC information, however, would be a much more dramatic change from conventional procedures.

The initial NASA data link experiment with transport pilots consisted of a realtime simulation study that compared voice communications with data link messages for ATC communications and prerecorded ATIS (Automatic Terminal Information Service) reports. The results indicated that pilots liked using a data link system for routine ATC communications, although they still preferred voice communications for urgent messages. Many of the test subjects also voiced some concern about the amount of "headdown" time required for reading and typing data link messages and the possibility of losing the situation awareness they obtained from listening to the communications between controllers and other pilots in the vicinity. [Ref 7-21]

In order to explore possible solutions to these concerns, the Langley researchers began another simulation study in 1990 that incorporated several additional features. In addition to the written messages printed on the CRT screen, the study used a digitized voice to annunciate data linked ATC messages to the flight crew. Preliminary results indicated that the addition of the digitized voice made the twoway data link system much more acceptable to pilots. [Ref 7-22]

Although the simulation results were promising, researchers felt it was also important to test the data link concept in a realistic ATC flight environment. Flight tests could validate the simulation results and point out potential operational problems with the technology that might not show up under simulated conditions. In addition, the ATOPS researchers had learned that in the cost and riskconscious world of airline operations, flying a new transport technology in an actual transport class airplane was often an important step toward winning acceptance for the concept among manufacturers, airlines and the FAA.

The main question the flight experiments were designed to answer was not whether the data link technology would work, but whether the system would be acceptable to pilots. Consequently, one of the most important features of the experimental equipment was the cockpit display the pilots would use to interact with the system. The data link information was displayed on a CRT screen, and the Langley engineers designed the display format to be as easy to use as possible. The screen was covered with a clear, touchsensitive panel that allowed pilots to select choices simply by touching the appropriate place on the display screen, instead of having to use a separate alphanumeric key pad. The display also used a "windows" format similar to the type used in many personal computers, with three different layers.

The top layer, or window, was the smallest and always had visual priority over the other windows. It was used to display information sent from the ground to the airplane. The middle window was for the crew to compose messages to send to the ground, and the bottom, largest layer displayed the main menu page and the weather menu page. The top layer displayed was always the active window and had a white border and white print, while the inactive windows had blue colored windows and print. Each window had touchsensitive areas that were drawn as buttons. The pilots would touch the button areas to make choices, scroll through a message, or select different menus or information. When the pilot touched the screen, magenta touchtarget lines centered on the selection the computer thought the pilot was making would appear, and the button would turn from white to green. After the pilot lifted his finger, the button would then flash twice to indicate the command had been accepted.

The main menu contained six basic options, labelled ATC, Weather Menu, ATIS, NASA Ground, View Clearance, and View Messages. The bottom portion of the main menu was used to notify the pilots if they had nontimecritical messages from the ground waiting to be viewed. When the pilot touched a message waiting button, the message would be displayed in a small window. When the pilot had responded to the message, it would disappear. By touching the "Weather Menu" button, the pilots could access a wide variety of data linked weather information. Touching the "ATC" or "NASA Ground" button opened up the middle layer window, so the pilots could compose messages to data link to ATC or NASA facilities on the ground. The ATIS (Automatic Terminal Information Service) button opened up a menu that allowed the pilots to choose and receive the most current recorded weather, runway and other operational information for different airports. Selecting the "View Clearance" or "View Messages" buttons would present the pilots with a printed record of the clearances or messages transmitted and received during the flight to that point.

If a timecritical ATC message was received at any point, however, it would be displayed on top of whatever window was currently in use, and a digitized voice would repeat the message aloud to the crew. The flight crew had to respond to any timecritical messages before any other message could be composed or received. The flight crew had three possible options for responding to a timecritical ATC message. They could select a button that said "Unable," if they could not comply with the ATC instructions, or they could touch a "Roger" button to simply acknowledge the message and communicate their intention to follow the instructions. The third option was to select a "Roger/Enter" button. By touching that button, not only would a "Roger" message be data linked down to ATC personnel, but the ATC instructions would be automatically entered into the flight management system. With the touch of a single button, therefore, ATCrequested changes in routing, altitude, speed or heading could be acknowledged, accepted, and executed.

The ground equipment for the experiments consisted of a phone modem link to a digital, real time weather data base, a data link processor, two separate VHF data link sources that communicated with data link receiver/transmitters on the airplane, and numerous voice frequencies. Since the point of the flight tests was to evaluate pilots' acceptance of a data link system for ATC communications, the flights were conducted in the vicinity of the Wallops Island Flight Facility, where actual communications with ATC were not required. For most of the flight tests, NASA researchers on the ground acted as air traffic controllers, composing instructions, clearances and route modifications to send to the flight crew. When weather conditions forced the crew to fly under actual instrument conditions one day, the research pilots were still able to use the data link system. Messages between the pilots and the FAA controllers were simply relayed through researchers in the NASA ground station, who data linked the controllers' instructions to the research crew and verbally relayed the crew's data linked responses and requests to the controllers.

The NASA researchers wanted to evaluate how operational flight crews would accept the use of a data link system, so they selected pilots from five different commercial airline companies to fly the research flights. A total of seven twoperson crews were chosen and given one day of orientation in the TSRV 737 simulator before flying the data link experiments. Because the training was limited to one day, the researchers selected pilots who had previous experience using electronic flight displays and flight management systems. Each crew flew three different 250 mile circle routes that each included a takeoff, climb, cruise, descent, and approach to landing, so the flights could incorporate a range of typical airline/ATC communications. Each flight also included at least one route modification. In one of the circuits, the pilots used only voice communications with the ground controllers in order to establish a baseline against which the data link performance could be measured. The second and third circuits used data link as the primary communication source. Clearances for taxi, takeoff and landings, however, were always made with voice communications.

The errors, miscommunications and message repeats each crew experienced were tracked during the flights, and the crews were all debriefed after their flights were completed. The results were striking. There were five instances of confusion with voice communication, and none using data link. The pilots had to ask for messages to be repeated 46 times with voice communication. Messages only had to be repeated 12 times during the flights relying primarily on data link, and all of those repetitions were during portions of the flight where voice communications were being used. While there were seven errors with both voice and data link communications, all but one of the errors on flights using data link involved tuning the voice communication radio to the incorrect frequency. In other words, the ability to read, store and reread the data link messages effectively eliminated errors due to miscommunications and the need for the repetition of ATC messages.

The airline pilots conducting the flights agreed that the use of data link reduced their workload and allowed them to distribute it better, especially if ATC commands could be acknowledged and put into the flight management computer with the touch of a single button. Six of the seven crews also thought that data link would make an acceptable primary ATC communication medium for most flight segments, as long as voice communication was available as a backup. However, the crews all thought that communications in terminal areas should still be conducted primarily through voice radio to reduce the amount of "headdown" time in busy traffic areas. They also suggested that at least the timecritical ATC messages should be displayed in the pilot's forward field of view. [Ref 7-23]

The NASA experiments were the first flight tests using data link as a primary source for ATC communications. But the airlines already had a strong interest in the technology and were eager to get Langley's test results. The research flights using airline pilots were completed in early May 1990. Before a month had gone by, the human factors/data link group of the Air Transport Association had asked for a flight demonstration of the technology. Typical attendance for the group meetings was reportedly about 15 people, but 60 airline representatives showed up at Langley to see the data link system in operation. [Ref 7-24]

The Langley engineers also realized from the start that a data link system would require the support and involvement of the FAA as well as the airlines and airframe manufacturers. So although the flight tests did not evaluate the impact of data link on controller workload or ATC system safety, the researchers took several outside observers along on the flights, including an air traffic controller from the ATC data link design group. The controller's response to the data link system tested by Langley was extremely positive, and he thought that controllers would like to use data link for ATC airborne information with the exception of landing clearances. [Ref 7-25]

The FAA and the airlines were both interested in data link because they believed it could enable more efficient air traffic operations. The airlines, for example, saw data link as a way to enhance the efficiency of transoceanic flights. With only VHF and High Frequency (HF) radio communications, it was difficult for flights to maintain contact with ATC or the airline dispatchers far from shore. Consequently, it was difficult for flight paths to be changed en route, even if the pilots encountered different winds than were predicted. With a data link system, airline dispatchers could send updated wind information to their pilots via satellite, and the pilots could compute a more efficient flight path. The pilots could then send a rerouting request via satellite to a ground station in the U. S., and then through a VHF data link to air traffic controllers. If data link were combined with a global positioning system (GPS), the position of airplanes could be automatically sent to the oceanic ATC controllers so that they could view all the airplanes in their area as if they had radar. Armed with this information, controllers could allow more direct routing and closer spacing. Based on some preliminary trials, a data link system was estimated to have the potential of saving between 3,000 6,000 lbs. of fuel and eight minutes of flight time on a flight between Los Angeles, California, and Sydney, Australia. [Ref 7-26]

The FAA saw the use of data link primarily as a way of reducing frequency congestion and controller workload. However, a data link system might allow future enhancements to the planned CTAS procedures. Specifically, a VHF data link would enable the CTAS computer and each airplane's flight management computer to communicate directly. As a result, the computers could actually negotiate a safe descent profile optimized not only for that general type of aircraft, but for that particular airplane, taking into account its fuel, weight, and system performance. [Ref 7-27]

In 1993, FAA plans called for an ATC data link system for transoceanic flights by 1995 or 1996, with data link capability at key domestic en route, terminal and tower locations by late 1996. Although some questions still remained about what transmission format the data link system should use, the most likely option appeared to be a composite structure that used VHF frequencies, Mode S discrete code transponders and satellite links. [Ref 7-28]

As opposed to many kinds of new technology, data link did not have to be sold to the airlines. They wanted it as soon as possible and were even "pushing the FAA harder than the FAA could accommodate developing and implementing it." [Ref 7-29] As a result, manufacturers began incorporating data link capability in many new transport airplanes. Boeing, for example, included twoway data link in an updated suite of navigation and communication functions available for new or existing 747400s and new production 767s. The company also planned to include twoway ATC data link capability in its new B777 airplane. [Ref 7-30]

Data link technology clearly had potential for making flight operations in the ATC system more efficient. An equally important factor in the rapid transfer and acceptance of the research information by industry, however, was the timing of the Langley experiments. The NASA data link flights were done at a time when the airlines were already pushing the FAA to develop a data link capability for ATC communications. The successful flight tests lent the concept the exact kind of support and credibility the airlines were seeking, which is why they jumped at the technology so quickly. At the same time, the NASA research flights provided the FAA with a measure of confidence in the ability of a data link system to handle ATC communications as well as the acceptability of such a system to the pilots who would have to use it.