A conglomeration of senseless ideas that might eventually turn into something useful.
There is a particular company that is becoming increasingly concerned with the intersection of the aerospace and security industries. There has always been some acknowledgement that the two are intertwined, if not completely in their sense of purpose, at least in the sense of national defense. Projects of direct missile deflection and detection based on machine learning are already in the works (they can't exactly be tested, due to costs and...you know...possibly killing people and further poisoning the world, but it's not like that ever stopped anyone), and there are other smaller scale defense-based research projects going on through different state agencies and private companies. However, the largest intersection is probably in the offensive.
This has made me re-think the attack surface.
It is much easier to break things than it is to make them, so why not build a defense from the same attack infrastructure? At the very least, if a barrier cannot be created, what can be done to stop an attack (both virtual and physical)?
I know very little about the aerospace industry: building 3D models for NASA is as far as I got (I am horrible at physics), so my commentary on this will be very limited compared to someone who works at JPL or SpaceX or something...
Anyway.
Besides the obvious risks of warfare (missiles, spy satellites, whatever), the aerospace industry is also faced with more idle issues caused unmanned aircraft and 'faulty' equipment. 'Faulty' because things can be made to be faulty on purpose, but let's ignore that for now.
Among the obvious issues of unmanned aircraft, we have the common drones sent to patrol the border area of the United States and Mexico (is this considered aerospace? not sure...but it can be scaled). As proven by (I suspect) many people online and in conferences like Def Con, these can be hacked easily (apparently the cartels thought so, too). Military-grade drones would hopefully be more secure, but the same idea is present (to the bigger geeks out there: yes, I know they operate on an entirely different grid just like government issued laptops, but let me keep it simple for the sake of literally everyone else).
Among the less common issues that might be encountered are radar misguidance -for example, driving an enemy to deploy weapons on a digitally fabricated target, in effect lowering ammunition and displaying possible strength of attack - and satellite manipulation... but this is drifting more into general security issues than aerospace specific, so let me change heading.
Okay. This seems to be a better fit.
Back to the radars. Apparently, according to one of my professors (who definitely knows his shit), some radar systems out there are vulnerable to overflow attacks, and some people have designed a proof of concept (PoC) that targets this vulnerability to change the output of the system on the interface. This means that you might see aircraft that isn't really there or have many copies of the same detection. I'm probably messing up the explanation on one thing or another, but that's the general idea. In reality, it's the machine instructions that are changing, not the actual data, but for our sake, we can treat them as the same since they have the same impact on the person who is looking at the information output (or machine, I doubt it would make a difference, although a human would probably be better at telling that : "no, really there aren't, in fact, 200 giant air crafts surrounding you. Wth." )
As for satellite manipulation...that gets really shady really quick; and considering I'm very likely going to China in a few weeks (hint hint), I might want to shut up about this one, haha.
The thing that aerospace agencies are unlikely to be directly worried about are the things that seem to be unimportant. Think of the embedded Huawei espionage, for example. Who would think to check the devices coming straight from the source? Was it that the next in the chain completely trusted those before them or that they thought them incapable of bugging every single device? In any case, the issue was mostly that of lack of interest than anything else. In a sense, this can be expanded to issues of national security and aerospace. For example, who would think to pay attention to the on-board printer (FUNTENNA!!!) or maybe even a watch.
How can this all be used on the defensive side?
I honestly hadn't thought of this. It's very far from my usual area of interest, but the only two other options proposed by the company were blockchain based security and machine learning. So now I am thinking of this. A lot.
How can a human be modified to detect attacks as if they were happening on their own person? Maybe that's a bit of a stretch, but if we have direction-sensing implants, what's the limit?
