User avatar
pathetic piss kitten @kitten@elizabeth.cat
9mo
so, my current doubts stem from:

from the people behind OpenHaystack there is this
paper. in section 6.1 they explain:
Equation (1) derives a new symmetric key from the last used symmetric key with 32 bytes length. Equation (2) derives the so-called “anti-tracking” keys ui and vi from the new symmetric key with a length of 36 bytes each. Finally, Eqs. (3) and (4) create the advertisement key pair via EC point multiplication using the anti-tracking keys and the master beacon key d0
now, what confuses me is, looking at OpenHaystack's code, we can see that what they transfer to the device on flashing is a symmetric key (SK0 in the equations) and what they call a "public key"

but I'm under the impression that it's actually the private key, and thus the device needs to hold both the symmetric key and the private key (d0)
1
0
2
0

User avatar
pathetic piss kitten @kitten@elizabeth.cat
9mo
feels a bit... not as safe? anyone who got physical access to my tag could dump the firmware super easily and get both those values (and then leave the tag as it was so it kept tracking). it'd be nicer if the algorithm could derive p(i) from p(0) or p(i-1) instead
1
0
1
0
User avatar
pathetic piss kitten @kitten@elizabeth.cat
9mo
(my knowledge in cryptography is very basic, any insight from people more into it is welcome)
0
0
0
0