5,964 reputation
1927
bio website github.com/CodesInChaos
location Munich, Germany
age
visits member for 1 year, 10 months
seen 38 mins ago
stats profile views 109

Apr
29
comment Counter Mode: static IV but different keys
For secure network protocols you typically run an ephemeral key-exchange at the beginning, so you end up with a unique master key in any case. So using nonce=0 is actually a bit easier, but generating a random nonce is very easy as well.
Apr
29
comment Secure order preserving hash function
Read the CryptoDB paper. I think what they use is the best thing we have, but it obviously has significant weaknesses inherent in order preserving encryption.
Apr
29
comment Secure order preserving hash function
That can't really be a hash anymore. Your condition can only be satisfied by order preserving encryption. It implies having different results for each input, and not just that collisions are hard to find.
Apr
28
comment standard way to authenticate parties with public keys without certificates
You need some way to associate Bob with his public key. Either per finger print verification (SSH style) which shifts to problem to obtaining the fingerprint, per trusted third party(CA), per web of trust,... You alway need some mechanism. Else how do you define "Bob", if not by his public key.
Apr
28
comment Adding and multiplication in jacobian coordinates
Take a look at the Explicit-Formulas Database
Apr
28
comment Adding and multiplication in jacobian coordinates
what's the curve equation? And what do you mean by multiplying points? I only know point addition and doubling and multiplying a point by a scalar by repeated addition/doubling.
Apr
27
comment Secure encrypt-then-sign with RSA
@RickyDemer No, I mean sign, then authenticated symmetric encryption together with RSA encryption on the symmetric key. The authenticated part of that is mainly to prevent annoyances that could break the confidentiality. (padding oracles etc.)
Apr
27
comment Secure encrypt-then-sign with RSA
I prefer sign then authenticated encrypt. The only tricky part is including enough information in the signed part - it needs to be able to stand for itself.
Apr
27
comment implementing long term archive encryption
There is no asymmetric encryption/keyexchange that I'd trust for 30 years.
Apr
26
comment How can a key pair be derived from an arbitrary hash?
Factorization only applies to RSA, not to ECC. For ECDSA a private key is just a number in a certain range. Numbers outside of that range can be mapped into the range with the modulo operator.
Apr
25
comment Password verification
Don't read more than the natural size from PBKDF2. Since Rfc2898DeriveBytes uses SHA-1, reading more than 20 bytes is a bad idea. It halves your KDF performance, without hindering the attacker. For client side password hashing I'd also crank up the iterations a bit to 100k or even higher. If that makes it too slow, replace .net's implementation with a fast implementation. Or switch to scrypt.
Apr
24
comment Using the output of a stream cipher, how to guarantee the integrity of 4 bytes of data?
"Concretely use the stream as a key for SipHash" what do you want to say with that?
Apr
24
comment Create groups of keys
This sounds like a program recommendation question, which is off-topic here. An appropriate question here would be how a protocol for accomplishing this would look like, but I suspect that's not what you're interested in.
Apr
23
comment Using the output of a stream cipher, how to guarantee the integrity of 4 bytes of data?
Are you sure that the overhead actually matters? You could also use a MAC with small constant overhead, such as SipHash.
Apr
23
comment Using the same secret key for encryption and authentication in a Encrypt-then-MAC scheme
I recommend HKDF as the KDF when you want to derive multiple keys (for example encryption and MAC) from a single high-entropy master key(not a password).
Apr
22
comment How is CipherCloud doing homomorphic encryption?
@adrenalion The letter, has two parts. One part is the DMCA notice for the images. The other (non DMCA part) is about the false, misleading and defaming statements.
Apr
20
comment A lower bound on the insecurity of CipherCloud?
They claim that their encryption is not deterministic and defeats frequency analysis. Which is a really weird claim and doesn't fit the encryption shown in Sid's screenshot.
Apr
17
comment Why is $h(H, m) = E(m, H) \oplus m$ insecure?
What exactly is the definition of $H$ and $E$? I guess $H$ is the IV or chaining value, and $E$ some form of encryption, but which? In particular which parameter is the key, and which the message? And is it AES, a PRP, or an ideal blockcipher?
Apr
16
comment Is it safe to assume Salsa20 to be a PRP?
Salsa20 takes 16 constant bytes for a reason. While it's build from a permutation, that permutation has some limitations and it has no key.
Apr
16
comment Is it safe to assume Salsa20 to be a PRP?
The random function is produced by xoring a permutation with with its input, similar to the way block ciphers are turned into hashes.