Changing leakage model in CPA attack

Hello, We are currently running PA_CPA_1-Using_CW-Analyzer_for_CPA_Attack.ipynb on chipwhisperer lite, when we changed the leakage model we aren’t able to retrieve the key for the unprotected AES scheme. Our question is, in the code why are we changing the leakage model after capturing the traces aren’t we supposed to do that before? Are there any other parts of the code to change the leakage model? We aim to attack aes rp (https://github.com/coron/htable/blob/master/src/aes_rp.c), we are trying to attack the output of the SBox of the last round (this was suggested in this paper https://eprint.iacr.org/2022/327.pdf) So, we wanted to change the leakage model.

No, capturing traces is independent of the leakage model, unless the attack requires the input plaintext to have specific characteristics, such as for example a MixColumn attack: https://github.com/newaetech/chipwhisperer-jupyter/blob/master/courses/sca201/Lab%202_3%20-%20Attacking%20Across%20MixColumns.ipynb

No.

Jean-Pierre

Thanks for the reply, but why aren’t we able to retrieve the key by changing the leakage model on the unprotected AES scheme?