ADS-B devil is in the details
Automatic dependent surveillance broadcast (ADS-B) is poised to make the transition from promising technology to fundamental air traffic management tool, and the trials helping prepare the way are identifying many of the details that will need to be addressed.
Both the future European air traffic system being defined by the Sesar program and the U.S. Next Generation Air Traffic System (NGATS) expect to rely on ADS-B for surveillance, with mandates for airborne carriage of the equipment likely to emerge in the next couple of years. The technique’s principal attraction is the cost of the ground stations, which is at least an order of magnitude less than that of a conventional radar, without even counting maintenance and other running costs. That in turn would translate into a reduction of 0.5 to 1 percent in air traffic control en route charges, a compelling argument for the airlines.
But if radars are suddenly to be regarded as legacy systems, their replacement must be able to demonstrate at least the same level of safety. That may not be difficult in low-traffic areas, but its applicability in busy airspace such as the London terminal control area needs further investigation. So the UK National Air Traffic Services (NATS) is working with Qinetiq (Hall 4 Stand C12), Raytheon (Hall 2 Stand A12) and SITA on the CRISTAL UK program to determine the role of passive surveillance in the future system. This will cover both ADS-B, which relies on GPS-derived position information broadcast by aircraft, and wide area multilateration (WAM), which correlates the reception of transponder replies at multiple sites to calculate an airplane’s position.
A first phase report compared the 2004 radar surveillance system, which is known to be safe, with a 2015 scenario involving ADS-B used in a mosaic system with radar. In the mosaic system one of the four radars covering the TMA is the default radar, the second is the stand-by radar in case the first becomes unavailable, and the other two provide first and second contingency back-up. One focus was whether the increased use of separations resulting from traffic growth could be traded off against the increased accuracy provided by ADS-B.
The study established that ADS-B ground stations could provide coverage of all terminal control airspace for southeast England at and above FL 55 (about 5,500 feet). It found that track update rates equal to or better than those available with radar could be achieved at a range of 100 nm, and possibly greater, depending on the ground system. And it showed that the trade-off of increased traffic against increased accuracy was feasible, so the future ADS-B scenario would have a lower collision risk than the radar scenario.
However, some items provided by mode-S enhanced surveillance in the radar scenario, including the mode-3/A call sign which is needed for correlation with legacy system information, will not be available through the current Airbus implementation of ADS-B. That means the controller would have different data items displayed depending on whether he used radar or ADS-B, which is undesirable.
It also found that although the ADS-B scenario has more hazards than radar, they can be mitigated to ensure safety is maintained. The key risk is small, credible corruptions of data going undetected, which could be mitigated by an alerting tool to compare ADS-B to a radar or multilateration-derived position.
Recommendations from the first phase included the use of a multi-sensor tracker to provide a composite radar/ADS-B display for presentation to the controller. This would overcome some of the inconsistencies in data presentation that could occur between radar and ADS-B and might also allow surveillance to continue if an aircraft was not equipped with ADS-B.
In addition, the use of multilateration as an independent means of surveillance for backup or verification of ADS-B should be considered as a means to identify credible errors in ADS-B position.
The second phase of the evaluation, which has 50 percent funding from Eurocontrol’s Cascade program as one of the CRISTAL validation projects, is looking at live ADS-B data collected by Raytheon ground stations at Heathrow, Gatwick, Luton and Stansted airports, linked by the SITA network and analyzed using Qinetiq’s processing and analysis facility at Malvern.
The key questions for NATS are whether passive surveillance will provide better surveillance and whether the company will be able to extend the area in which it provides a three-nautical-mile separation service by supplementing radar operating at ranges where it supports only 10-mile separation with ADS-B.
More than 2,000 aircraft have been seen broadcasting ADS-B data in the TMA, and more than 50 percent of aircraft equipped with mode-S are ADS-B capable, but there are variations in both the data broadcast and the quality of the data. Of 620 aircraft identified during one five-hour session, for example, only 379 were broadcasting call sign, airborne position and velocity, though all but 41 broadcast call sign and velocity.
One measure of the data quality is provided by the navigational uncertainty category (NUC) figure, which indicates the airborne system’s own estimate of the accuracy of its position. An unexpected result of this can be apparent abrupt changes in position as the NUC value improves: in one case tracked by NATS, the target appeared to move back 1.19 nm and jump 0.16 nm laterally as its NUC figure improved and the accuracy of its position was refined.