Earlier this year, in Anchorage, Alaska, a Federal Aviation Administration (FAA)-sponsored international conference on advanced aviation technologies heard speakers from the United States, Australia, Sweden and Russia. They described advancements under way in their countries and elsewhere to implement automatic dependent surveillance-broadcast (ADS-B) as part of an International Civil Aviation Organization (ICAO)-endorsed worldwide process towards enhanced flight safety, improved surveillance and increased airspace capacity.
Technical specialists from ICAO, Eurocontrol, RTCA and airframe and avionics manufacturers also presented information, as did members of the user community, from general aviation to airfreight operators and commercial airlines. The overall view was that ADS-B implementation was moving ahead despite certain technical questions that remain unanswered.
The Largest Program
By far the world's largest ADS-B program to date is the FAA's Capstone initiative in Alaska. It is designed to provide "near-National Airspace System (NAS) quality" services to small commercial aircraft flying at low altitudes in remote areas, far beyond the reach of standard VHF communications and well below the coverage of air traffic control (ATC) radar. Capstone I started in 2000 around Bethel, in western Alaska, and today has more than 200 participating aircraft.
The Anchorage conference marked the commencement of Capstone II in the Juneau/Ketchikan region in southeast Alaska. It will involve a further 200 aircraft. Capstone I already has been hailed a success. Once the part of Alaska with the highest aircraft accident rate -- in a state that already had a much higher accident rate than any of the other 49 states -- the Bethel region has seen a 40 percent reduction in accidents. It now is the Alaska region with the lowest accident rate.
FAA has several other initiatives under way. Installation of a continuous network of 23 ADS-B ground stations is in progress along the U.S. East Coast, and completion is slated for 2005. The agency also plans to establish ADS-B "pockets of implementation" at locations where system usage is expected to be high. The first of these is at Prescott, Ariz., where Embry-Riddle Aeronautical University will shortly commence avionics installations in its fleet of 40 training aircraft. Sixty ADS-B-equipped aircraft at the university's Daytona Beach, Fla., campus will use the new East Coast network.
Embry-Riddle's program also underscores the diversity of ADS-B applications. The university's prime reason for adopting ADS-B is to reduce the probability of near mid-air collisions by allowing students to see other air traffic. By contrast, overnight freight operator United Parcel Service (UPS), the nation's ADS-B pioneer, wants the system to help pilots "self separate" their aircraft. The goal is to close up large gaps between planes arriving at the carrier's cargo hubs during the late night/early morning "rush hours," thereby increasing turnaround efficiency. Towards this end, the company will permanently assign up to 20 of its Boeing 757 and 767 freighters to be the last aircraft arriving each night at the Louisville, Ky., hub, in order to evaluate various ADS-B techniques under ATC radar monitoring.
In yet a different application, FAA has been testing ADS-B's ability to monitor aircraft flying across the central part of the Gulf of Mexico, which lies just beyond the coverage of U.S. and Mexican coastal radars. Here aircraft have had to revert to wide "procedural" track separations, which reduces traffic throughput and creates bottlenecks. However, ADS-B can provide air traffic controllers accurate "pseudo-radar" aircraft position monitoring. And if the trials prove to be successful, this could significantly reduce track separation standards and increase capacity in the center of the Gulf.
Australia has a somewhat similar, but much larger, problem. The country is only slightly smaller in size than the 48 contiguous U.S. states, yet much less densely populated. Between the string of ATC radars along Australia's east coast and the few, isolated radars in the west, procedural separation is the rule. However, installing and maintaining a U.S.-like, nationwide radar network would involve an "astronomical" cost, especially in the country's barren interior, where many installations would have to be unmanned and autonomous, according to Greg Dunstone of Airservices Australia (ASA) at the Anchorage conference.
ASA therefore has decided to implement ADS-B as a pseudo-radar "gap filler" to cover the country's interior. The agency has ordered 28 ground stations from Thales ATM and plans initial operating capability in 2005. When complete, this will expand high-level (30,000 feet and above) surveillance capability from less than 20 percent of the Australian continent to more than 99 percent and allow radar-like separation standards. In a nod to the FAA, Dunstone stated that "Australia's ADS-B activities would not have progressed as far without Capstone leading the way."
However, as a privatized ATC service provider that generates its operating funds from user fees, ASA had to first obtain approval of its ADS-B plan from its "customers," including foreign air carriers, since the required investment would mean increased user fees. The air carriers agreed. They even went one step further by approving in their fee structure the cost of subsidizing the installation of equipment in Australia's 8,000 to 9,000 general aviation (GA) aircraft, to increase overall safety.
Dunstone also is chair of ICAO's Asia/Pacific ADS-B Implementation Task Force. This body is developing an industry-wide cost/benefit study and a follow-on regional implementation plan, targeting a 2006 start date. In Dunstone's words, "Things are now hotting up," and most nations represented on the task force have commenced preliminary work.
ADS-B is advancing in Europe, as well. Eurocontrol has established a multinational requirement focus group (RFG) to drive its CRISTAL (Co-opeRative Validation of Surveillance Techniques and AppLications) program. The RFG will "work towards implementation of ADS-B in Europe with global interoperability" and "bring the first set of ground and airborne surveillance applications and infrastructure, enabled by ADS-B, into reality in Europe." While no implementation time frames were given at Anchorage, the Europeans clearly do not intend to be left behind in the international transition to ADS-B.
Separately in Europe, Sweden started its evaluation of ADS-B as far back as 1996, and has been active in many Eurocontrol test projects since that time. The Swedish Civil Aviation Administration (CAA) already has drawn up plans for nationwide implementation of ADS-B by 2007 and later this year will launch a 12-month GA program in high-density terminal airspace. It will involve 50 aircraft, "from gliders to twin engines," according to a CAA official.
Russia faces a challenge like Australia's. It has a vast land mass, much of which is remote and sparsely populated. Air traffic density is high on Russia's western and eastern sides but much lighter in its central and northern regions. Russia therefore plans to install ADS-B ground station networks only in the high-traffic density areas; it will employ data links via satellite communications systems in its lower-density airspace under an approach analogous to the ADS-C (for contract) used today by airlines crossing the Pacific Ocean.
Russia participated in earlier ADS-B trials with Sweden, France and Italy, and intends to launch a major pre-operational implementation project in its western region beginning this year. It will cover five airports and involve two domestic airlines. Officials expect that more than 90 percent of the nation's aircraft will be ADS-B- or ADS-C-equipped within 15 years. But they will press ICAO to declare an eventual worldwide installation mandate in every aircraft and airport vehicle in the interest of safety.
While virtually all nations agree on the benefits that ADS-B will bring to aviation, they disagree on which data link technology should be used to transmit the signals between aircraft and between the aircraft and the ground stations. Three link techniques have been proposed: the Mode S transponder-based 1090ES (1090-MHz frequency with extended squitter, the signal bursts that carry the ADS-B data); the Universal Access Transponder (UAT); and the VHF Data Link Mode 4 (VDL-4). Each link is distinct from the others. They are technically incompatible, and each has strong advocates. Further, none is clearly superior to the others, causing the November 2003, ICAO Air Navigation Conference to carefully conclude that, "each of the three data links could be seen to have relative advantages over the others with regard to some criteria." The conference did recognize that 1090ES was a common element in most approaches adopted for early implementation of ADS-B, but endorsed the continued development of technical standards for UAT and VDL-4.
The strongest supporters for 1090ES are the major air carriers. Virtually all of them use Mode S transponders, which they soon will upgrade to meet new Eurocontrol elementary and enhanced surveillance requirements. The upgrade automatically includes integration of the 1090ES capability. As a consequence, the United States, Eurocontrol, Australia and the Asia Pacific nations have all adopted 1090ES as their future data link medium.
So why UAT and VDL-4? The 978-MHz UAT band is, in fact, an FAA-sponsored development, aimed at providing a small, lightweight, low-cost airborne ADS-B transceiver specifically for the Capstone project but also for general aviation at large. UAT is the exclusive data link in Capstone; and, as FAA implements its ADS-B ground station network across the United States, the agency will install a mix of UAT and 1090ES sites to accommodate both airline and general aviation users. Meanwhile, Airservices Australia seeks a lower-cost (no more than $3,500) Mode S transponder for its GA aircraft operators.
Recently, researchers have developed a signal diplexer that allows the UAT signals to be transmitted and received via the standard Mode S transponder antenna installed on most transport aircraft. However, while FAA had hoped that last year's ICAO conference would support international implementation of UAT, this was not to be, and its use probably now will remain limited to North America.
VDL-4 was developed in Sweden in the early 1990s and has successfully demonstrated its potential over a wide range of ATC applications beyond ADS-B. Like the UAT, VDL-4 is a smaller, lower-cost alternative to 1090ES. And like the FAA, Sweden also proposed that ICAO should endorse its international adoption. This also was not to be.
Operating in the 108- to 137-MHz frequency band, VDL-4 is far from the UAT and 1090ES part of the radio spectrum, so separate antennas would be required on Mode S-equipped aircraft. This would no doubt make VDL-4 less appealing to the air carrier community.
But Sweden is not alone in supporting VDL-4. Finland and Russia have adopted the concept, and Elena Gromova of the Russian State Research Institute of Aviation Systems (GosNIIAS) told the Anchorage conference that her country's planned ADS-B implementation would use the VDL-4 data link. Gromova also stated that in several ways, VDL-4 was further advanced than 1090ES as an ADS-B data link. She pointed out that in high-traffic density areas, 1090ES could potentially become overloaded. Western attendees at the conference acknowledged this, and it appeared to be generally accepted that a completely new ADS-B data link would be needed by 2015, and preferably well before that date. In the meantime, however, VDL-4, like the UAT, probably will be limited to a regional role.
Another issue that arose at the conference concerned "holes" in GPS receiver autonomous integrity monitor (RAIM) performance. Current GPS receivers, built to TSO C-129 specifications, use the RAIM technique to check the integrity of the system's satellites. They do this by comparing the geographic fix positions from different combinations of any four of the five or more satellites in good "geometry" above the local horizon. If all satellites are operating correctly, the positions derived from any combination of any four of them should be the same. If one satellite is not performing properly, the receiver's calculated positions will be different, and the pilot will receive a RAIM alert, warning that the system should not be used for navigation. A RAIM alert also will be given when there are just four or fewer satellites in good geometry. This can happen because the satellites are constantly moving relative to each other and the user.
How often does a RAIM alert happen? And how long does a RAIM "hole" or outage last? Australia's Dunstone reported that monitoring records over two and a half years at his Bundaberg (Queensland) ADS-B ground station site showed that RAIM outages occurred on average every 10 days and averaged 19 minutes. The individual duration could last from one or two minutes to more than 40 minutes.
However, because the satellites' future movements can be determined precisely, the location and duration of these outages can be predicted. Nevertheless, this has important implications for ATC, since ADS-B-equipped aircraft transmit their positions based on GPS, and an aircraft's effective loss of navigation impacts its separation assurance from other traffic. In turn, this could require controllers to revert to wider and less efficient procedural separation standards in a non-radar environment, which happen to exist throughout most of the world's airspace.
A solution proposed at Anchorage was to replace all "legacy" C-129 GPS receivers used in ADS-B applications with newer units built to the TSO C-145 specification. These units do not use RAIM, but instead use integrity data from the GPS wide area augmentation system (WAAS) or the European geostationary overlay system (EGNOS) or other international equivalent. None of these augmented systems normally suffers from such integrity outages. However, some concern was expressed at the Anchorage conference that aircraft operators might resist a requirement to purchase newer, and more expensive, GPS receivers.
Despite these technical issues -- which undoubtedly will be resolved -- there remains a clear consensus that ADS-B will become a key element -- some say the foundation -- of the world's future air traffic management scheme. Enhanced safety, improved surveillance, increased capacity, more efficient routes and many other benefits are predicted from the technology.
A Broad Overview
Aircraft equipped for automatic dependent surveillance-broadcast (ADS-B) carry special transceivers. Once per second, they transmit the aircraft's GPS position, identification, course, speed, altitude and intent (climbing, flying level or descending) over a common radio data link. All other ADS-B equipped aircraft within reception range receive the data, which then appears on their cockpit displays, giving pilots a clear picture of other aircraft in their area.
The aircraft transmissions also are received by widespread unmanned ground stations, which retransmit them to the nearest air route traffic control center (ARTCC). This allows controllers to track distant or low-altitude aircraft flying beyond or below the ARTCC's radar and integrate them into the overall traffic picture. At the Anchorage ARTCC, controllers routinely monitor the movements of ADS-B-equipped bush planes flying at low altitudes in western Alaska, several hundred miles away and well below radar coverage.
In turn, the ARTCC can send flight and traffic information back to the ground stations, which then retransmit it to local aircraft as flight information service and traffic information service broadcasts (FIS-B, TIS-B). FIS-B provides pilots with weather reports and forecasts, notices to airmen (NOTAMs), pilot reports and other important flight data, while TIS-B automatically adds the positions and movements of non-ADS-B-equipped aircraft to the pilots' cockpit displays.
In the transition to ADS-B, when equipped and non-equipped aircraft are flying in the same airspace, TIS-B will provide a valuable safety benefit.