As the GNSS world starts to appreciate the era of multiple global constellations, it’s probably worth considering the impact on aircraft navigation, GNSS airborne receivers and what these changes might soon bring to those who develop and use GNSS for airborne en-route navigation and approaches. Design and manufacture of new breeds of receivers are only the first of many steps along the road to full-fledged use of new capabilities. In aviation in particular, and in other high-precision fields to a lesser extent, many stages of study, regulation, development, test, and certification must be undertaken to eventually reach the promised land. The following capsule history of our progress to date also sheds light on what we may have to undertake in the future.
When GPS first came along, it was seen as a godsend for aviation. We’d been struggling with long-range Omega, which gave us, wow, close to a mile accuracy, and trying to get affordable but highly accurate Dopplers out of the military world. And those expensive inertials which still drifted pretty fast. GPS let us get to meters of position accuracy, was affordable, and could be used together with inertials to give us both bounded long-term accuracy and inertial azimuth, elevation and roll. Autopilots loved this stuff!
But, hey, when you put electronics onto an aircraft, you need standards and you need regulations. The International Civil Aviation Organization (ICAO), to which virtually all air-faring nations subscribe, sets up international standards for airborne system performance. ICAO quickly understood that GPS would be the navigation system that the whole world would come to rely on. So we got top-level marching orders from ICAO that set out the basic performance that we should expect from an airborne GPS system — Minimum Aviation System Performance Standards (MASPS).
The Federal Aviation Administration (FAA) in the U.S. as the world’s leading aviation agency took up the challenge and turned to the Radio Technical Commission for Aeronautics (a volunteer organization that develops technical guidance for use by government regulatory authorities and by industry — RTCA) to put together the Minimum Operational Performance Standards (MOPS). MOPS provide standards for specific equipment(s) used by designers, manufacturers, installers and users of the equipment. In fact, when we got MOPS for GPS, it allowed manufacturers of receivers and avionics to figure out what to build, what it should do, and how to qualify it. MOPS allowed the FAA to move on to develop the Technical Standard Order (TSO), which allowed FAA staff to verify compliance and ultimately install GPS avionics on aircraft via a Supplemental Type Certificate (STC) or, for new aircraft, via a Type Certificate. There are other ways onto aircraft too, but these are probably the most commonly used.
So we don’t just write our own specs and just develop and test it and sell it to aviation — there are a host of regulations you say…? That’s correct.
Around the same time GPS was hitting the streets, the FAA and other agencies recognized that these gizmos were basically software widgits, and that nerds could even be developing them in their garages or on their kitchen tables. We wrote software then that got the job done, and it worked at least as well as Microsoft’s did at the time, but our design and testing might have benefited from more structure. So along came RTCA DO-178 and subsequent revisions that set out the sequence of steps you should take to ensure a low probability of undetected error in your software. We started spending a whole lot more time upfront figuring out what the requirements were before we designed anything, and we had to wait what seemed like eons to get to code the stuff. So although we got process and we were able to cross-check between each step that we were still doing what we set out to do, the level and intensity of what came to be known as “verification” went through the roof. What a couple of guys in jeans had previously been able to cobble together in a couple of months now began to take a group of ten people working as an exceptionally well-coordinated team, maybe over several years. And the bright creative guys in the garages and kitchen tables were out of the aviation business. And a good proportion of the software engineers in the industry actually began to shun the rigor of the airborne software qualification/certification process, and prefer the less regimented receiver development associated with commercial receivers, especially challenging complex dual-frequency and RTK applications. So engineers with the persistence of airborne software qualification specialists are a rare and desired breed for airborne receiver manufacturers.
And then FAA put the same processes in place for hardware, especially ASICS, and gave us RTCA DO-254. Luckily we were already implementing ASICS using tools that applied thoroughness and intensity to the process, so DO-254 didn’t hurt quite as much as software development changes had hurt.
But we persevered, and eventually we got the bugs out of the receivers and got people flying with them, and there was great rejoicing in the aviation community. The General Aviation (GA) guys were a little impatient and couldn’t wait for the lengthy aviation development wheel to turn, so they bought handhelds in the meantime and duct-taped them to their yokes. But the FAA caught up with them, and these receivers evolved to the point where today they are buried behind square feet of glass Electronic Flight Instrument Systems (EFIS) and all you get is an icon which says your navigation source is GPS.
All these GPS receivers were, of course, single-frequency L1 only. Why, when there are perfectly good high-precision dual-frequency L1/L2 receivers out there? Well, turns out that only L1 is within the protected Aeronautical Radio Navigation Service (ARNS) band. The International Telecommunications Union (ITU) is an agency of the United Nations that manages worldwide allocation and use of radio frequencies — every country subscribes and plays by the ITU rules. So L1 is protected for GPS navigation use, but L2 isn’t. You can blast away with X-band radars in the L2 band and you’ll interfere with GPS L2 quite effectively. LightSquared assaults on the GPS bands excepted, the regulations currently protect you and your L1 GPS aviation receiver from intended and unintended jamming while your flight management system using GPS flies you safely from LaGuardia to Boston Logan.
So after several million dollars of engineering sweat and tears by receiver manufacturers, we eventually got ourselves to a point where we could fly GPS enroute from place to place, and the FAA even constructed and approved GPS approaches so we can get into airports. Then along comes WAAS, and we have new MOPS and new TSOs and many more millions of R&D required to implement WAAS LPV (Localizer Performance with Vertical guidance) approaches into the already certified L1 GPS receiver population. Crank up the machine and ensure each step is verified and the checks and balances of the aviation certification process ensure that we have the safety level needed to fly plane-loads of people through clouds and down to 200 feet with a half-mile visibility to the runway. It’s taken a lot of people a long time to work up these safety levels to use GPS like this, and guess what — it works!
And Automatic Dependent Surveillance – Broadcast (ADS-B) is rolling out across the U.S. and other nations as we use GPS on-board aircraft to tell us where they are and to allow agencies to monitor air traffic with less radar tracking. Another leap in safety levels, which requires airborne receivers qualified to new MOPS.
Even though the accuracy might be there, the element we’ve struggled with all this time is integrity. And we get integrity by adding signals and ensuring there are always more signals than we have currently. So now we have L5 coming, and even Galileo and the prospect of multi-frequency, multi-constellation airborne receivers that add safety, raise integrity and get us closer to the runway on approach.
The FAA is mulling over the ground rules for the new L5 MOPS, which they would like to have done and receivers ready for when the next generation of WAAS is ready to go. Over in Europe, Eurocae (the European equivalent to RTCA in the U.S.) has already started to think about Galileo, and even L5. Over the coming years, as the U.S. GPS constellation adds L5 capability, maybe even at the same rate that Europe puts Galileo in place, aviation receiver manufacturers — a rare breed of specialists indeed — are trying to figure out how to finance the impending investments needed to make and certify this new generation of aviation receivers.
Some people have already started down this torturous development path, so they will be ready as the L5 and Galileo MOPS committees work through the intricacies of adding capability to the already highly structured spider’s web of regulations and requirements. As concepts are thought out, it’s always great to be able to test if they work on prototype hardware and provide validation back into the requirements process, so those who have working receivers will play a key role as these new capabilities are brought on line.
And let’s assume that Europe will want Galileo in its future aviation system, and that the pace of fielding L5 into the GPS constellation is unlikely to pick up given the economic restrictions under which the DoD will have to operate in the next few years.
So, combined GPS L1/L5 Galileo E1/E5a receivers are favorite candidates for the next generation of airborne receivers. Rockwell Collins, Honeywell, CMC, Septentrio (at right, the Septentrio AiRx2 L1/L5 GPS E1/E5a airborne receiver), Accord, Garmin, and potentially others have their work cut out as we move out into a new generation of aviation satellite navigation capability. And yes, it’s very hard to do, folks.