April 8, 2013

Space Based Fallacy

Perhaps some of you are familiar with "The Praxis Foundation".  If not, I invite you to meander your way over to one of my adjacent sectors on the right column.  You may note the seemingly paradoxical subject of their article about ADS-B and the required ground stations required for what is being sold to the public as a more efficient space-based airplane tracking system.  The new trend in aviation technology development (aka:  NextGen) is to fund the heck out of idea and then wait around a decade or two for the results.  The hardest part is not being disappointed by the end.

Disappointment is inevitable, however.  The companies who bid for the initial development funding are naturally going to promise the world to us on a silver platter.  They probably even intend to deliver what they promise.  But I'm not here to talk about promises.  I'm here to talk about expectations and results.  I actually don't even want to talk about expectations, but I have no choice than to bring them up to frame the subsequent, inevitable disappointments.

ADS-B - What I expected and how reality stacks up:

Let me start by saying that most controllers, myself included, do not see radar as a limitation to capacity.    Now that I write that, I bet you couldn't find a controller who would think that.  ADS-B is in use in Louisville, KY with UPS, but I don't know anywhere who works there....  Ok, let me rephrase:  I don't think you can find a Center controller who thinks radar is a limitation to capacity.

In Boston Center, we have a significant portion of our airspace, below FL180, certified for reduced lateral (3NM instead of 5NM required) separation. A lot of work goes into getting 3NM airspace.  It gets justified by saying "but we can run planes that much closer, which means we can run more in the same amount of space!".  Three miles is great when you're on a small range running a bunch of planes in a straight line (like, say, an approach control).  In my area, the scope is generally configured to be 200 miles from one side of the scope to the other.  The computer symbol that displays the location of the airplane is about 1 mile wide.  So you want me to run 3 miles apart!?  At 450 knots ground speed!?  Nobody does that intentionally.  Very often, the wind at one end of my sector will be significantly different speed and from a different direction than the other end.  And its changing at the speed of the wind (50-200 mph).  There needs to be a buffer so when one plane enters the area of more headwind and slows down, he doesn't get run over by the plane still goes much faster right behind.  The point is, we don't need less than 5NM in the en-route environment.  So, I don't expect ADS-B to help me increase capacity.  I am obligated to bring up the point that the MAIN factor in increasing capacity and reducing delays is adding runways that we can land on every day.

I won't just dismiss this as a waste of time, however.  As a controller who isn't completely under-layed by approach controls (I "own to the ground") I am acutely aware of the limitation of my radar and where I have no coverage at low altitudes.  So, when I heard about "space-based tracking" (ADS-B), I was excited that one day I'd be able to see all of my planes all of the time.  I expected that such a space-based system would look down over the world, send a signal down to the planes, receive a reply back, and then the satellite would beam down the data to our scopes.  It would be able to see everyone with an operating transponder.   This assumed system would not be able to fully replace radar, since planes in distress sometimes loose their transponders in some form.  We wouldn't be able to assist planes when they needed our help the most.

But, that isn't how it currently works.....

As of right now, I am getting mixed messages.  Boston Center ran a test a few months ago to see how our ADS-B coverage compared to our normal radar coverage.  ADS-B is required to have coverage that is the same or better as radar.  But ADS-B is not certified to be used in places we currently don't have radar coverage or within areas we use 3NM.  It is also not designed to see places radar can't.  If we show an improvement, that is strictly convenient.  Currently, ADS-B is used in Alaska and over the Gulf of Mexico (these places don't have good radar coverage).  If you are confused, join the club.  But what about those damn satellites!?  Can't they see everyone!?

Well, my expectation don't meet the reality.  Did you read the Praxis article about all of those ground based sensors?  The "space-based" part isn't anything other than good ole' GPS that the airplanes use to navigate today.  So, how it works:

Step 1 - The plane's GPS figures out where it is and converts that data into coordinates.

Step 2 - The new, super-awesome, transponder in the plane takes those coordinates, throws in pressure altitude data, aircraft ID info, and maybe even some other data that ATC isn't going to use, and, every second, beams it down to a ground based sensor.   Think cell phone tower........

Step 3 - That data collected by the ground based sensor goes off to a server somewhere and distributes the data to places that need it.  The aforementioned altitude, location and ID data ends up in a an FAA computer which then sends it to our scopes.  (This last step may be slightly simplified)

When you drive your car to the airport using a GPS on your phone or dashboard, you are equally, if not more "NextGen", than the flight you're about to board.  Luckily, you are much less likely to get lost since there are actual people making sure you don't make a wrong turn.  These people also help you avoid traffic jams and accidents.  Its very convenient.

Safety - Why I'm not writing it off yet....

We already have a network of ground sensors in place.  They exist.  We used them in February for a test.  They work.  So, how can we use them to make the system safer?

First, planes with ADS-B will be able to see other ADS-B equipped planes around them.  Callsigns and everything.  The ground sensors can transmit traffic information (TIS-B) out to planes to display on their own screens.  If your plane is out of range of a sensor, some systems will be able to receive the data via relay through other nearby airplanes that are in a location or at an altitude that is within range of a sensor.  Some sensors will also include FIS-B, which would transmit new NOTAMs, AIRMETS and/or SIGMETS for the area.

The push now is to give controllers the ability use high altitude planes to relay information to and from planes out of radar and sensor coverage.  The concern from the developers is that all of this relaying back and forth would take up too much bandwidth.  That's fine.  Just give us an ERAM function to toggle the function on and off.  We just need one or two at a time to fill in the gaps.  We don't need all the planes to be relaying back and forth to all the other planes.  We get it.  Just let us switch between one, and then maybe another when we need it.  If the next space-based system is supposed to razzle and dazzle us, then it should be an improvement!  The next system should make us wonder how we ever lived with just radar.  And I say "just radar" because you can't ever turn radar off if you want to maintain safety.  And that's a controllers primary concern.  Safety.  We are supposed to be there when you  need us.  We need to be able to see you if you want our help.  It's our job to think of reasons ADS-B can't be the only way or else someone who doesn't care about safety is going to make it the only way.

Reasons we can't get rid of radar:

1 - Terrorists turn off the transponder.
2 - Plane looses electrical power
3 - Attenuation.
4 - For that matter, a huge tornado takes out some sensors....
5 - ADS-B antennae get covered in five inches of rime ice.
6 - Lightning?
7 - Power outage at ground sensor
8 - Hackers
9 - Solar Flares

Got more?  Leave a comment.

Reasons why ADS-B increases safety of flight:

1 - Increases pilot awareness.
2 - Allows controllers to provide expanded traffic separation and advisory services.
3 - Is a great backup to radar.
4 - Is very accurate over a larger area (radar gets less exact the farther away from the antennae you go).

Got more?  Leave a comment.

As a taxpayer, perhaps your first question should be: "why didn't we design a system which met the goals of the second list, with an emphasis on safety of flight?"  Surely that would be much less expensive and the project would have had a clear goal.  Can't help ya there, sorry.  I'm busy working planes; not enough time to alter the universe.

I will add here, before I go, that ADS-B will enable future programs to safely integrate ways to increase efficiency.  Wow, I should get a job as a NextGen salesmen......  Seriously, though - there is testing in the works that will allow planes to safely run closer together and/or to sequence themselves in a fuel efficient manner using the data that is relayed back and forth between the planes as I mentioned above (Imagine the Perfect Vector, where planes always went exactly 5 miles behind another when crossing over SYR.  Or, using speed, planes approach the active runway perfectly staggered on opposite base legs, ready for the final controller to put them 3 miles apart on final.)  Imagine it.  Some controllers would feel threatened.  Others would imagine the economy recovering at some point and the task of working twice as many planes as we do now.  We're gonna need some help.....

Till next time....


PS.  The airlines were totally on board with this program until they realized that the government wasn't going to buy the new fancy transponders and traffic screens for them.  Suddenly, it wasn't such a great money saving piece of NextGen technology.   Planes equipped with ADS-B are few and far between at this point.  ERAM envisions a world where almost all planes have ADS-B.  When a plane doesn't have it installed or operating, a little red A appears on our scope next to the plane.  Luckily, we can turn them off, since they are everywhere.  So, ERAM can dream....but it is still learning how to make a handoff to New York approach.  :)  Due to furlough and supposedly resolved issues with NY which delayed it, ERAM at Boston Center is postponed until after the end of the year, or when furlough ends.  Or whenever.....

PPS. Oh, by the way.  ADS-B stands for Automatic Dependent Surveillance - Broadcast.  If you asked.


kilwer said...

My terminal radar facility has recently transitioned to using the ADS-B technology known as fusion. Most of the recent technological advancements over the past several years at our facility have truly improved the way we do our jobs. Fusion, on the other hand, has been a major disappointment.

A phenomenon known as "stitching" causes radar targets to jump around in random directions. This is due to multiple sensors attempting to display the true position of an aircraft. It seems as though the stitching problem becomes more critical as the aircraft require turns to the final.

After observing the "true" position of aircraft for over 30 years using primary ASR, it is a downgrade of technology to use fused radar.

One can only hope that the advantages that pilots enjoy using ADS-B outweigh the degradation of radar presentation that terminal radar controllers must endure.

Anonymous said...

I've always wandered about the same potential issues that you listed. It just doesn't sit well to think that the people responsible for safety of everyone "up there", are unable to track someone, without them "flicking the switch" first.

One interesting point to note, is that a lot of places outside of the US seem to wholly depend on SSR/MSSR radars for surveillance. It's hard to say whether each country's military plays some role in still keeping an eye out for iffy primary radar returns, but the AIPs sure seem well groomed in stating what equipment is available. What's even more interesting about the wording, is that fairly often, it'll describe the primary radars as not being a usual piece of equipment, in the provision of services...even when available.

Perhaps someone who actually works as a controller in such places can confirm it, but the official publications sure clearly give the impression described above.

CenterPuke88 said...

ZFW ERAM grinds to a halt on April 21. We trained all but the last 10% of the workforce before the furlough was official and we were directed to stop training. Now we hope that we will restart in time to just do a mild refresher for most people, train up the last few, and off we go.

Of course, if this grinds on for any amount of time without restoring funding for ERAM, we'll be starting all over again...and all that training and overtime goes down the crapper.

For those reading this and thinking "Hey, that furlough was over-blown. Nothing has happened in ATC yet, it was all a scare tactic!", just wait until about April 22nd. The first ATC furloughs happen April 21, but will be light since that's a Sunday and the beginning of the pay period...by April 29th (at latest), expect to see some real impact.

XR650L_Dave said...

Comments on the article, and on the comments, in no particular order.

Fusion refers to taking the data from multiple positioning sensors, weighing the accuracy and update rate etc of each, and coming up with the best estimate of the position. Fusion certainly should not lead to targets jumping around, that sounds much more like mosaiced radar. Sounds like on a per-region or per-update the surveillence decides to use a radar track or the ads-b data. Someone did a sales job on that one...

NAV CANADA is going to be playing around with space-based ADS-B, no ground stations, I forget the name of the company- aireon or such?

I believe DoD/DHS is going to be keeping enough spinny primary radars so that the 'homeland security function' can be maintained relative to border security.

For areas beyond the cares of border security, I think that they are pondering using the new (ground based of course) phased-array weather radars to also keep tabs on un-cooperative aircraft (electrical failure, no electrical system, disabled transponder, etc).

The reason they are planning on so many ground stations is so that coverage will be much better than with existing radar, down to lower altitudes. Ground station placement will decide how true this is.

I more often mull other aspects of nextgen I find troublesome, such as how this critical national-asset infrastructure will be owned and run by a private company (ONLY duty is to the shareholders), and how this public-fund generated data will then be available to only those who can afford to pay for it- ITT gets paid by the FAA to generate the data, then ITT also gets to sell the data.


Amado said...

This is cool!