Sunday, May 17, 2009

The Sky is Failing


If you read this blog very often you've heard me caution pilots about being overly reliant on any single method of navigation, like GPS. Since the GPS was declared fully operational in 1995, pilots, sailors, hikers, drivers, cell phone users, and the military all have become accustomed to the convenience provided by this no-cost system (okay, you have to buy the GPS receiver). Even the next generation air traffic control system NexGen, which the FAA has been furiously flogging as the panacea for everything from air traffic delays to tooth decay, relies heavily on GPS. But those plans might need to change.

A recently released GAO report entitled Global Positioning System: Significant Challenges in Sustaining and Upgrading Widely Used Capabilities has many wondering about the stability and sustainability of our constellation of GPS satellites, whether or not the Air Force can improve and "replenish" the system in time, and what our daily lives on the ground, at sea, and in the air might look like if GPS became unavailable or degraded.

The GAO's 61 page report is interesting reading, especially when compared to the glowing claims made by the FAA about the progress that's been made enhancing GPS availability and growing the number of RNAV procedures for aircraft. And if you are wondering if GPS is an aging system in need of repair or a robust system that continues to be expanded and enhanced, the truth probably lies somewhere in between.

One key questions that needs to be addressed is this: What is the useful service life of a GPS satellite? A solid answer is hard to come by, but one thing of which we can be certain of is that satellites have worn out in the past and more will wear out in the future. Of the current constellation of 31 GPS satellites, thirteen entered service between 1990 and 1997, twelve entered service between 1997 and 2004, and the remaining six satellites were launched between 2004 and 2009. That means about a third of the GPS satellites are between 12 and 19 years old; a sobering thought.

One of many problems that aging GPS satellites can experience involves clock errors. Since GPS receiver's position is calculated based on signals from several satellites, knowing the time each of the signals was sent and the time it was received is crucial. Each GPS satellite has four clocks that are periodically refreshed from the master station on the ground and each satellite will broadcast an estimate of time offset of the onboard atomic clock from the GPS system time. Since GPS signals travel at the speed of light, even minute errors in clock settings or offset estimates can result in large positional errors for GPS receivers. According to the GNSS Evolutionary Architecture Study released in February of 2008:
... large clock runoffs were experienced on SV22 [space vehicle #22] on July 28, 2001; SV27 on May 26, 2003; SV35 on June 11, 2003, and SV23 on January 1, 2004. These events generated range measurement errors of 1000 meters or more . The pseudorange error on SV22 on July 28, 2001, was reported to be 200,000m by some users and 300,000m by others.

Other problems that can afflict aging GPS satellite's include the failure of positioning components, propulsion systems, and a degradation of the power supplied by each satellite's solar array. The GAO report points out that the Air Force can often switch to a satellite's back-up system. And they can manage the power problem to an extent by powering down satellites when they are not needed or by shutting down power to "secondary payload" systems. These sorts of measures may extend the service life of older satellites, but it's clear that nothing lasts forever. New replacement satellites must be planned for and launched on schedule.

This leads to the GAO's main concern with the US Air Force's poor track record of creating and launching new satellites on schedule and within budget. Not only has the track record in Phase IIF been poor, the schedule for Phase IIIA actually compresses the lead time to produce and launch satellites. Without going into all the details of feature creep, government contractor inefficiencies, and the like, the bottom line is that the Air Force may not succeed and we GPS users may have to do with fewer satellites. Here's the GAO's estimate of how many satellites we may have to do with, or without. If they are right, the probability of having 24 functioning GPS satellites between 2010 and 2012 hovers around 85%.


The FAA's enhancement plans for GPS follows several phases and the FAA's aggressive creation of RNAV approaches with LPV minima in Phase II has been nothing short of astounding. Only a few years ago I was writing about LPV approaches from a strictly theoretical standpoint and now they outnumber ILS approaches. We are now in Phase III and a few of the stated goals are to :
  • Provide LPV approaches with 200' DA (most are currently 250 feet)
  • Transition support and enhancement of WAAS to the FAA
  • And prepare for an increase in predicted solar activity
The FAA appears to be doing a good job and enhancements to the WAAS ground components have significantly expanded WAAS RNP 0.3 availability.




And LPV coverage has been extended significantly, too.



Increasing GPS receiver accuracy using the Space-Based Augmentation System (WAAS) is obviously on track. People who study such things predict that there will be an increase in solar ionospheric disturbances beginning somewhere around 2022 and lasting for some number of years. Solar activity has reduced WAAS availability in the past and it's reasonable to expect it will happen again. Some have suggested that LORAN could be a back-up system, but for aviation that idea seems dead on arrival: There are few functioning LORAN units out there and no new units that I know of currently being manufactured for aviation use. The FAA has decommissioned many NDBs and they plan to decommission more (basically not fix the ones that fail). They'd also like to do away with most VOR stations. For better or worse, we seem to have put all our navigational eggs into the GPS space basket.

And the future could be very interesting indeed.

1 comment:

Trevor said...

I would think that a NDB would cost a whole lot less to maintain then and VOR station. It's just a radio antenna right? Also I would think that would be a better back up for GPS because of it's lone range expect for the fact that that is hasn't been a hot peace of avionics for a while. but if the FAA is taking down both NDB and VOR I'd like to see the NDB stick around.