We have a product under development that uses AES-128 with a Master Key XOR'd with the device's serial number and a single-use randomly generated IV to encrypt a password before transmission. Passwords are fixed length.

I have verified that an attacker can capture the serial number, IV, and cyphertext-password during a connection attempt. If the attacker buys our product they can also set the password. So they will then have the serial number, IVs, and as many plaintext/cyphertext-password pairs as they want.

Assuming they can put it all together, is this enough information to reverse the Master Key? It seems to be like it is, but I'm a little out of my element here.

I believe my question differs from the possible duplicate because in this case the attacker can generate unlimited amounts of fully-known plaintext/cyphertext. In particular, I believe that our device meets the requirements for vulnerability to power-analysis side-channel attacks.


1 Answer 1


Knowledge of the plain-text and it's direct encrypted cipher-text does not offer the attacker any method of reversing the master key other than brute force, which they will already have at their disposal by just knowing the cipher-text. Therefore, your system should still be secure.

To find a lot more detail and justification to this answer, I would recommend this post: How less secure is an encryption if we know something about the original data?

Hope it helps!

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .