An extremely well attended talk by Hugo Teso, a security consultant at
n.runs AG in Germany, about the completely realistic scenario of plane
hijacking via a simple Android app has galvanized the crowd attending
the Hack In The Box Conference in Amsterdam today.
Teso, who has been working in IT for the last eleven years and has been a
trained commercial pilot for a year longer than that, has combined his
two interests in order to bring to light the sorry state of security of
aviation computer systems and communication protocols.
By taking advantage of two new technologies for the discovery,
information gathering and exploitation phases of the attack, and by
creating an exploit framework (SIMON) and an Android app (PlaneSploit)
that delivers attack messages to the airplanes' Flight Management
Systems (computer unit + control display unit), he demonstrated the
terrifying ability to take complete control of aircrafts by making
virtual planes "dance to his tune."
One of the two technologies he abused is the Automatic Dependent
Surveillance-Broadcast (ADS-B), which sends information about each
aircraft (identification, current position, altitude, and so on) through
an on-board transmitter to air traffic controllers, and allows
aircrafts equipped with the technology to receive flight, traffic and
weather information about other aircrafts currently in the air in their
vicinity.
The other one is the Aircraft Communications Addressing and Reporting
System (ACARS), which is used to exchange messages between aircrafts and
air traffic controllers via radio or satellite, as well as to
automatically deliver information about each flight phase to the latter.
Both of these technologies are massively insecure and are susceptible to
a number of passive and active attacks. Teso misused the ADS-B to
select targets, and the ACARS to gather information about the onboard
computer as well as to exploit its vulnerabilities by delivering spoofed
malicious messages that affect the "behavior" of the plane.
Based on his own research, Teso developed the SIMON framework that is
deliberately made only to work in a virtual environment and cannot be
used on real-life aircrafts. His testing laboratory consists of a series
of software and hardware products, but the connection and communication
methods, as well as ways of exploitation, are absolutely the same as
they would be in an actual real-world scenario.
Since it's nearly impossible to detect the framework once deployed on
the Flight Management System, there is no need to disguise it like a
rootkit. By using SIMON, the attacker can upload a specific payload to
the remote FSM, upload flight plans, detailed commands or even custom
plugins that could be developed for the framework.
To make things even more interesting - or easier - Teso showcased an
Andorid application that uses SIMON's powers to remotely control
airplanes on the move. The application, fittingly named PlaneSploit,
sports a clean and simple interface, but is packed full with features.
This is a remarkable example of technology evolution - ten years ago we
barely had phones with a color screen, today we can use them to hack
aircrafts.
PlaneSploit uses the Flightradar24 live flight tracker and you can tap
on any airplane found in range. When talking about the range, please
keep in mind that we are talking about a proof-of-concept application
used in a virtual environment. In real life, the range would be limited
depending on the antennas used (if going directly for the plane), or
global (if misusing one of the two big ACARS players such as SITA or
ARINC).
The user interface is divided by its main functions which are
self-explanatory: discovery, information gathering, exploitation and
post exploitation. The attacker can click on any active airplane and is
receives its identification, current location and final destination. In
case a nearby airplane system is exploitable (a number of vulnerability
vectors mentioned, not much details provided), the application alerts
the user via an in-application alert or a push message. The payload can
be uploaded with a tap of a button and from that point on, the flight
management system is remotely controlled by an attacker. There are a
number of other systems connected to FMS, so further exploitation is
possible.
Here are some of the functions Teso showed to the HITBSecConf Amsterdam audience:
- Please go here: A way of interacting with the plane where the user can dynamically tap locations on the map and change the plane's course.
- Define area:
Set detailed filters related to the airplane, for example activate
something when a plane is in the area of X kilometers or when it starts
flying on a predefined altitude.
- Visit ground: Crash the airplane.
- Kiss off: Remove itself from the system.
- Be punckish: A theatric way of alerting the pilots that something is seriously wrong - lights start flashing and alarms start buzzing.
By
showing a sample scenario of a drunk pilot flying over Berlin, Teso
mentioned that the Android application also uses the benefits of the
accelerometer and therefore a remote attacker can transform the motion
of its smartphone into physical changes in the plane's movement.
It's amazing to discover that aviation - an industry where safety is of
vital importance and every physical element has one or even two
fail-safe mechanisms - is failing to secure the onboard computer, the
heart and brain of the plane.
Teso has not shared too many details about the tools he used to effect
the attack, as the vulnerabilities have yet to be fixed. He says that he
was pleasantly surprised by the reaction of the industry to his
research and discoveries, as the companies didn't try to deny the
existence of the problems and have vowed to aid him in his research.
He says that older, legacy systems harking back to the 1970s will be
difficult, if not impossible, to fix, but that modern ones will easily
be updated with patched and modified firmware and software.
The vulnerabilities, of course, differ from system to system and from
plane to plane, but it's easy to discover just which ones are present
once the attacker identifies the type, model of the plane, and the
airline for which it flies.
There is a solution for pilots to regain the control of the plane and
land it safely, he says. Attacks of this kind work only when the
auto-pilot is on, so the trick is to switch it off, then fly the plane
by using analog instruments.
The bad news is that there aren't that many on modern planes, and that
the pilots have to detect that the plane's computer is being hacked in
order to effect these maneuvers, and that is no easy feat.
