U.S. and European aviation authorities struggle with dwindling communications bandwidth
By J.R Wilson
WASHINGTON — Sometime between 2007 and 2010, leaders in the United States and Europe agree they will exhaust the capacity of their current systems of allocating frequencies for air traffic control (ATC). When that happens, federal authorities either must put new systems in place or halt the 4 percent annual growth in air travel.
Part of blame for this problem lies with the U.S. Federal Aviation Administration (FAA) and the European Organisation for the Safety of Air Navigation — better known as Eurocontrol. These agencies took irreconcilably different routes in dealing with short-term frequency problems in 1995. At that time, officials of the International Civil Aviation Organization (ICAO) authorized two approaches — splitting the then-common 25 kHz separation for ATC-authorized frequencies down to 8.33 kHz and pursuing a new way to use 25 kHz called VDL-3 (VHF Data Link, Mode 3). The FAA program considering VDL-3 is called NEXCOM (Next-Generation Air/Ground Communications System).
Basically, VDL-3 uses Differential 8 Phase Shift Keying (D8PSK) and employs 4.8 kilobits per second voice code/encoder — better known as VOCODER — for voice operation to enable as many as four channels to operate on one 25 kHz frequency assignment. The most commonly supported format calls for a 2-Voice/2-Data configuration, although other combinations are possible. Where current aviation VHF radios use carrier sense multiple access (CSMA) — or push to talk — Mode 3 is based on time divided mode access (TDMA), where users are assigned a time slot and transmit within that time slot on that frequency.
Under the current system, civil aircraft operate in the VHF band between 117.975 and 137 MHz, while military aircraft operate in the UHF band between 225 and 400 MHz for air to ground communications using Double Sideband Amplitude Modulation (DSB-AM). Yet using 25 kHz channel separation allows only 760 assignable VHF channels in the U.S., with almost one-third of those designated for non-ATC aeronautical use. Different ATC centers that do not overlap and interfere with each other's transmissions can use most of those frequencies repeatedly across the continent. Aircraft operating at maximum altitudes, however, can each frequency only three times across the continental U.S.
There currently are more than 10,600 ATC frequencies assigned in the United States, with each controller having a separate frequency for each designated sector of airspace. Little has changed in the basic structure of this system since World War II, aside from segmenting frequencies to increase the number of channels. In the U.S., the last time that happened was 1972, when the system went from 50 kHz to 25 kHz spacing, while Europe opted for the 8.33 short-term solution in 1995.
Multimode radios with VDL-3 capability are on the drawing boards at Honeywell Commercial Electronic Systems in Redmond, Wash., and Rockwell Collins Air Transport Systems in Cedar Rapids, Iowa. However, deployment of them would be delayed by about a decade — two years of FAA rule-making, two years of development, a period of certification testing, and about five years for the airlines to install them during regular Schedule C maintenance rotations. This puts full fleet implementation out to about 2010, at the earliest.
A new VDL-2 (digital data only) radio from Rockwell Collins — the VHF-920 — was to take part in PETAL II (Preliminary EUROCONTROL Test of Air/ground data Link, Phase II) aboard four American Airlines 767-300ER jetliners in May. It was to be a multimode unit built on an existing 8.33/25 kHz box.
"We don't anticipate going from Mode 2 to Mode 3 being a complete radio replacement. We expect it to be an upgrade to the VHF-920," says Rich Ross, Rockwell Collins marketing manager for communications systems. "We expect right now to be starting that development in 2002, with a couple-of-year development cycle. We haven't put any cost estimates out on those upgrades yet; there's too much definition that needs to be done first."
Honeywell radio designers are working with the European Aeronautic, Defense and Space Company (EADS) in Munich, Germany, to certify thier RDA-44D VDL-2 radio on the Airbus A319/320/321 jetliner by March 2002 and on the A330/340 jetliner by September 2002.
"When Mode 3 comes along, we will upgrade the RDA-44D to handle digitized voice functions. For Mode 2, we are plugging a separate module into the existing (8.33/25 kHz) radio; for Mode 3, we expect it to be primarily a software upgrade, although there could be some minor hardware changes," says Karl Klewer, Honeywell's marketing manager for communications and navigation radios.
Both companies' plans reflect the U.S. approach to incorporate multimode avionics at 25 kHz, 8.33 kHz, VDL Mode 2, and VDL Mode 3 in one box.
"The radios the airlines now are buying already handle everything but Mode 3 and the plan is to develop avionics to have the radio handle that as well, so the air carriers would be compatible with both U.S. and European systems. That way, they can buy one radio and set it up in whichever mode they need it to operate," says Jim Eck, air/ground communications product team lead at the FAA.
"We are looking at full system validation and demonstration in 2004. We will have our key site implementation in '07 and full implementation by '09," Eck continues. "We're buying radios today that are multimode and doing the system development between now and '04, do the dem/val [demonstration/validation] in '04 and publish the rule shortly thereafter. It's a fairly large system. By '09 we will be installed in 20 centers and 1,500 remote radio sites, so the magnitude of what we are doing on the ground is fairly large and all aircraft flying above 18,000 feet will have to be equipped. When we cut over a system to the new waveform, only properly equipped aircraft will be able to fly there. So there is a lot of work to be done on both the air and ground sides."
European needsWhile having such multimode radios available for full fleet installation by the end of this decade meets the U.S. schedule, it does not answer the European dilemma."Because they can't recombine those 8.33 channels back into a 25 kHz channel so they can go to TDMA, they may have to look at alternate band solutions beyond the VHF band," says Don Willis, manager for spectrum planning in the FAA's international division.
Europe was forced to take the 8.33 route because they already were experiencing a severe frequency shortage in the most heavily trafficked sectors of the continent in the late 1990s and the digital solution that VDL-3 offered was nowhere near being ready. Nor could they have turned to a satellite-based system, although satellite communications (satcomm) are being used on trans-oceanic flight segments. The problem is time latency — as much as five seconds in getting a message from a ground-based controller to the aircraft. In the wide expanses over the ocean, that is rarely a critical concern, but for tactical communications in the crowded skies over major cities, it is not plausible.
Eurocontrol officials also are considering the aviation-protected Microwave Landing System (MLS) frequencies for communications. These frequencies have rarely been brought into use despite efforts by the FAA and Europe since the 1980s. FAA officials also are considering MLS frequencies for terminal use.
"The next step for Europe will be the wideband communications at 5 GHz, which is how they will fill the 20-year gap between about 2010 and 2030," says Jim Chadwick, director of communications, navigation and surveillance systems at the Mitre Corp. Center for Advanced Aviation Systems Development in McLean, Va. "And maybe at 2030 we all finally come back together with satcomm [satellite communications]. That technology has to mature and the operational concepts to use it have to mature. I don't see that happening within the next 10 years."
The upshot of these separate solutions means aircraft flying between the U.S. and Europe for at least the next generation will need multimode radios capable of handling a host of communications systems. About 25 percent of the U.S. fleet and virtually all of the European fleet already have 25 kHz and 8.33 radios; only part of Europe is using the 8.33 system, with the rest still using 25. The first multimode radios equipped with VDL-2 will begin going into the fleet this year.
But by 2010, commercial airliners also may have to accommodate VDL-3 for the U.S. and wideband — probably incorporating spread spectrum techniques to overcome power issues — for Europe.
"If you go from VHF to C-band, that's a frequency difference of about 50 times, which would require 2,500 times as much power," Chadwick explains. "There are ways to reduce that using spread spectrum."
Commercial airliners are not the only concern, however. In fact U.S. airspace has far more general aviation (GA) and business aircraft than airliners. Government officials must consider these aircraft, too, when they propose new communications equipment. Private aircraft traffic is beginning to blossom across Europe, as well.
The military factorYet another player in this mix is the military, which must operate in civilian airspace above both continents.The U.S. Department of Defense (DOD) "is a major factor in all these discussions," Chadwick says. "One thing we want to avoid is creating a situation where a new radio or navigation system implemented by the FAA ends up breaking the bank for the military. There are a lot of ways to mitigate that. A substantial portion of the DOD fleet only uses UHF anyway and those are unaffected. For those that do require VHF, either by the nature of their missions — flying in and out of civil airports or talking to other aircraft in the VHF band — it is necessary to make sure they are equipped. And in a lot of cases, they don't have the space for new boxes."
Concerns over lack of spare onboard space and cost also are of obvious importance to the GA and business aircraft communities.
While FAA and Eurocontrol leaders believe the measures they have in place will forestall serious problems until the end of this decade, not everyone agrees. "Anybody who thinks we can wait until 2009 for a change isn't talking to the right people," warns Tony Broderick, former FAA associate administrator for regulation and certification who now works as an aviation safety consultant. "We can't survive until 2009. Everyone I talk to says we need a lot of frequencies available and we need them relatively soon; we already need a lot more sectors today. You are able to handle more air traffic by generating less density in each sector. I don't see how a few adjustments here and there can do that.
"Indications I've heard say in a few years — well before 2009, say around 2005 — we will have had so much growth we won't be able to handle it with what we have today. This is a big problem that has the potential for stifling growth and in the short term being the source of significant increases in delays because the air traffic system won't have the ability to adapt to the growth in flights."
Broderick also is less than enthusiastic about the prospect of forcing everyone to install new multimode radios capable of communicating with a half dozen different systems: "How many multimode things do we need to have on airplanes before we get some sense on this and reach some international agreements? We haven't really been doing a very good job of that lately and its important to do so."
The first real step toward finding a U.S. solution — short- and mid-term — is a report that was scheduled to go to FAA Administrator Jane Garvey last April from the NEXCOM Aviation Rulemaking Committee (NARC). Created in December 2000, it consisted of representatives from industry — including the National Air Traffic Controllers Association (NATCA) — and did not include anyone from the FAA itself.
That report is not expected to be made public and Garvey may not act on its recommendations for several months. Moreover, aviation industry consensus probably will not settle on the two primary solutions in the report — VDL-3 or a combination of 8.33 and VDL-2, says committee chairman John Kern Sr., a retired Northwest Airlines executive who previously had served as Broderick's deputy at the FAA.
"We've had Europeans in to make presentations to us, as well as the avionics industry," Kern says. "We've tried to explore all the possibilities and heard all the arguments on all sides. We're trying not to leave any stone unturned — nor do anything shortsighted. I think the FAA intends to consider our report, then put out a notice for proposed rulemaking as expeditiously as possible — we hope within six months of our delivering the report. We hope one of the things our group will be able to do is take it out of the category of a controversial rule, which may facilitate the rulemaking timeframe. But who can predict rulemaking?"
While committee members are not allowed to discuss their deliberations or recommendations, the lack of consensus that Kern fears probably stems from strong support for VDL-3 by some groups and waning support from others.
"Mode 3 probably is a good idea, but inertia is against it at this point," notes Bill Stine, director of international operations for the National Business Aircraft Association (NBAA) and head of an RTCA infrastructure special committee that has been advising the FAA on frequency issues since 1991. "A sub-group of NARC is trying to get a handle on the economics and what might be best. Mode 3 is like any technology — not perfect. It is even less perfect when you are 10 years after it was first proposed and still several years away from it being reality."
Stine says he believes VDL-2 implementation will happen, but adds it is "not sufficiently deterministic for ATC; you can't be certain how fast a signal will go through electronically, computer to computer. TDMA can give you digitized voice and data and seemed like a reasonable compromise, but it took a decade to get to the MOPS [Minimum Operational Performance Standards].
"When we have had places where ACARS [Aircraft Communications And Reporting System] — sort of a primitive Mode 2 — has been used to download data, it has lightened up pre-departure clearances," Stine says. "At places like Dallas, it has greatly lightened the load on voice communications. That was a first step, but we have to take more first steps. The question is how much can Mode 2 realistically do? We know CSMA works, but how does TDMA work in real life?"
Stine also echoes Broderick's concern about proliferating systems, but from the perspective of business and general aviation: "If the U.S. does something different from Europe, then we have to put a box in the aircraft that is different to operate here than needed to operate there. So there is a groundswell for a homogenous system."
That "groundswell" does not extend to the air traffic controllers union.
"We had a meeting at the end of last year to once again discuss 8.33 versus VDL. Basically, my feeling is there is too much of their 8.33 camel in my VDL tent," NATCA safety and technology director Bill Blackmer, a member of NARC, says bluntly. "If we don't do VDL, we can't do datalink because the spectrum isn't wide enough."
Interestingly, Blackmer also raises a point that might be used against VDL: "We're talking about using GPS [the satellite-based Global Positioning System] for navigation and surveillance. The GPS timesource also is part of VDL. So if there is a GPS outage, then we can't see them, they can't see each other, they can't navigate, and we can't talk to them. That's bad. And that's why we have always argued for a secondary system, which does not have to be as robust. But from what I've seen of VDL technology, it is simple physics. The radios have to be made to high standards, naturally, but we already have people building radios that way and I know they can implement this new technology."
Whatever recommendation NARC makes to the FAA for mid-term relief (basically from 2010 to 2030), the challenge of meeting an expected 3 percent per annum growth in the demand for ATC frequencies in the U.S. still must be met between now and the end of this decade.
Toward that end, FAA officials are considering a list of some two dozen initiatives, most of which were put forth at the end of last year.
"These initiatives are more in terms of economizing where they can, finding available frequencies at both the top and the bottom of the band," says Bill Sears, program manager for air traffic technology applications for the Air Transport Association and a member of NARC. "They're going back to ATC facilities to see if they have a frequency here or there they aren't using. It's a general scrubbing of inventory to try to improve the efficiency."
Some of the initiatives being studied are to:
- investigate use of part of the 121.5 MHz guard band for ATC (121.5 is an internationally protected aviation distress frequency, but new radio technology has made guard bands reserved on either side of it redundant);
- review requirements for the current two flight-check frequencies;
- review the Federal Communications Commission (FCC) frequency utilization plan and investigate the use of UNICOM and other FCC aeronautical frequencies for ATC;
- investigate the possibility of using flight service station (FSS) channels for ATC, including the frequencies 123.6 MHz through 123.65 MHz;
- investigate utilization of the band 136-137 MHz;
- revise air/ground frequency assignment model to accommodate 0.6-nautical-mile vertical separation; and
- investigate combining HIWAS and ASOS/AWOS on VOR channels.
"We have to deal with a harsh reality — the U.S., Europe and Asia are all on completely different schedules with respect to the maturity of their infrastructures and the urgency of their demands," Chadwick says. "And to try to get all those synchronized is a tall order, to say the least."