Timeline for Can we use a Merkle Tree structure to reduce digest size?
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 16, 2020 at 11:26 | comment | added | mephisto | You can try to analyze this, but you should note that the adversary is not forced to keep DATA0 in your example. In any case, you can think about the new hash function that is obtained from your construction by fixing DATA0. Assuming that your starting hash function has good random behavior, this behaves like a 128 to 128 bit hash and you did not really win much (except the constant factor overhead of making a call to the new hash cost 2 calls to the initial hash). | |
| Dec 15, 2020 at 9:58 | comment | added | Modal Nest | Thank you for the response. I'm not being very clear (per usual) but the aim is to make 2nd/pre-image attacks more difficult. I.e. An attacker can find a pre-image, but the data making the correct hash also needs to "SPLIT&XOR" into the parent. I'm not sure if it adds much, but it can't be as simple as finding an unconstrained pre-image. On your 2nd point, I'm using "collision" when I should use "pre-image" in the question, I think. I'll try and edit. Obviously finding generic collisions doesn't attack a Merkle tree anyway. | |
| Dec 15, 2020 at 8:57 | history | answered | mephisto | CC BY-SA 4.0 |