Computers in Air Traffic Managment

The selected industry I’ve chosen is the air traffic management industry where computers play an incredibly important role – automation. 

A brief history

Historically, the bottleneck for national airspace access has been air traffic controllers. Early air traffic control was accomplished by the post office using signal fires, flags, and large painted arrows on the ground. Aircrews would fly relatively low so as to be able to see these navigational aids. 

As planes grew more complex, so too did the technology necessary to guide them. In the 1930s, navigational aids evolved in rotating lighted beacons. Air traffic controllers began operating over radios, controlling aircraft using time over fix, airspeed, estimated time over next fix, and other tools of the trade to guide aircraft without tracking their location via radar. Flight progress was tracked on chalkboards and relied heavily on the mental acuity of the controllers, but responsibility for safety of flight fell squarely in the laps of the pilots. 

Post WWII, ever-increasing air traffic congestion led to multiple midair collisions which had the public demanding radar installation throughout the country in the late 1950s. Further evolution saw aircraft transponding via beacons which provided secondary radar information to supplement the primary radar returns (which had previously been tracked with strips of paper on “shrimp boat” strip holders that air traffic controllers manually followed the primary radar returns on their scopes with).

Air traffic management systems continued to evolve to meet the increased demands that stressed the limited situational awareness of air traffic controllers. Automated air traffic management systems were developed which could recognize future conflicts, often hours in advance. The FAA has been attempting to continuously upgrade air traffic management automation since the 1970s, with mixed success. The first attempted project was such an abysmal failure that it is widely regarded as one of the most terribly managed project in project management history – it also led to the 1981 air traffic controller strike which saw President Reagan fire nearly every air traffic controller in the nation. 

Use of computers in Air Traffic Management

While individual use of personal computers in air traffic control is somewhat limited, systems continue to be developed which provide increased automation and enable controllers to handle greater workloads. It’s crucial that controllers continue to familiarize themselves with these systems and their inner workings in order to have a greater understanding of the limitations of these technologies. Unfortunately, most controllers have only a limited understanding of these systems because they’re so focused on keeping aircraft from colliding. 

More recently, technologies such as Traffic Collision Avoidance System (TCAS) and Automatic Dependent Surveillance-Broadcast (ADS-B) have been implemented. The former enables two equipped aircraft to detect potential collision hazard between themselves at a greater distance without relying so heavily on ATC or the pilots’ own MK-I eyeballs. The latter serves a similar purpose but was originally intended to be implemented as a replacement for radar coverage in regions where radar coverage wasn’t feasible, specifically the Caribbean. 

ADS-B

ADS-B highlights an important lack of security and privacy-mindedness regarding computers in the governments of the world. ADS-B has become a world-wide mandate despite numerous cybersecurity concerns. A little detail is necessary to explain just how bad the situation is. 

  1. First, ADS-B broadcasts aircraft identity, location details, airspeed, and more without any encryption.
  2. Second, these broadcasts are picked up by a terrestrial network of transceivers, many of which are privately owned.
  3. Third, no handshake or independent verification of the received information is possible – it’s quite simple to spoof an aircraft’s identity.
  4. Fourth, because the data is not encrypted and broadcast in real time (at 1Hz), ADS-B can actually be used to derive a targeting solution.
  5. Lastly, ADS-B has two major bandwidth issues:
    1. When message overlap occurs, the entire system becomes unreliable. Bandwidth combined with minimum transmission power make this more likely to occur, and in fact it has in the airspace over Florida, numerous times.
    2. There’s a user interface bandwidth issue. ADS-B displays on aircraft do not have an altitude filter which makes it nigh impossible to discern location data on potential threat aircraft when there’s significant congestion. Again, happens regularly over Florida. 

In fact, a white-hat hacker who goes by the handle RenderMan was able to teach himself how to inject a false ADS-B signal into the national airspace in just one weekend. He managed to do so responsibly,and managed to inject an aircraft with the callsign “YOURMOM” into the SFO Class B aerodrome and buzz the tower repeatedly (the controllers at SFO tower did not receive the transmission and there was no impact to flight safety). 


However, he gave up trying to convince people of the dangers of ADS-B after it was clear nobody would listen and has moved on to “the internet of dongs” and is advocating for a practical cybersecurity and privacy mentality as it regard IoT-enabled sex toys. 

Getting away from “the internet of dongs” and back to the ADS-B woes, this isn’t something limited to the United States, nor just to aviation. Aviation is a global logistics backbone. Consider that the economic impact of the drone incident at Gatwick was at least $124M. The absolute lack of cybersecurity mindedness with regards to the treatment of the national airspace as a network is both appalling and rampant. 

Recently, Boeing has been in hot water for espousing a short-term profit culture which prevented critical software risks from being mitigated – but Airbus will soon be eating crow. Airbus’s most modern helicopters and passenger aircraft have incorporated ADS-B collision alerts into their AUTOPILOT. Moreover, airlines (as a result of insurers guarding against human error) mandate autopilot while enroute, and in some case until 50 feet off the ground. So, any malicious actor could effectively shut down the next generation primary location information source for aircraft to prevent air traffic from doing their job, and inject false, non-verifiable signals which effectively steer airborne aircraft with up to 800 passengers on board. This is an systematic weakness which is being ignored and is a vulnerability that non-state actors could easily exploit to wreak economic havoc and is an asymmetric warfighting capability that the world has handed over on a silver, winged, publicly-broadcast platter. 

The current trajectory, no pun intended, of air traffic computer systems and networking is a move toward Four-Dimensional Trajectory Based Optimization (TBO) wherein aircraft are delayed on the ground for a couple of minutes to provide them with optimized routing hours later. 

However, new airspace entrants to include small unmanned aircraft and autonomous Urban Air Mobility aircraft (unmanned flying taxis) will throw a few wrenches into the works. Ultimately, privatization of unmanned air traffic management technologies will lead to the eventual replacement of both pilots and air traffic controllers in favor of automated systems with human-in-the-loop oversight. 

After all, the cause of most accidents is human error. 

However, that replacement is probably something like 30 years out. In the nearer-term (10 years) we will likely see an implementation of 4-D TBO and the start of the use of remote tower technologies to provide air traffic control services for terminal aerodromes without air traffic control towers, or that don’t operate 24/7. These remote tower technologies can also be used to augment the controller capabilities with infared optics, datablock overlays (instead of flight strips) and improvements in weather forecasting capabilities. 

The UAS realm will see the implementation of remote ID capabilities similar to those afforded by ADS-B (FAA indicates this is approximately 2 years out, so we can expect it in 3-4) but hopefully not ADS-B based. This will enable a greater scope of unmanned operations intermixed with manned aviation. As a result, the business case for manned aviation will slowly give way to unmanned as insurers come to recognize the increased risk of manned aviation. 

Hardware upgrades will be very slow, I recall having to change out 12-inch tape reels for out facility communications recorder and being excited that we were transitioning to cassette tapes – in 2009. FAA facilities will be upgraded sooner than USAF facilities but these technological paradigm shifts move at a glacial pace as a result of their governance by an insurmountably glacial Congress. 
Note: The views expressed here are my own and do not reflect the opinions of my employer or the USAF. All of the information discussed above is publicly available. 

0 Comments:

Post a Comment