I am trying LAB3_1 with chipwhisperer huksy (+ SAM4S in QFN48 Target Board).
At the end, after calculating the difference of the traces, I found an unexpected spike (marked with red circle) . What is this? (I found the same spike in SOLN).
I understand that the desired spike is the area marked with a blue circle.
These are also differences that result from the byte being set to 0x00 vs 0xFF. To understand exactly why would require analyzing the compiler output (which isn’t the purpose of this lab).
The exact shape of the differences will depend on the target and the compiler version; for example if you look at our pre-recorded traces (by setting VERSION = 'SIMULATED' in the notebook), you’ll get something else, which also shows more than a single area of significant power differences.
A key point here is that those significant differences are all occurring in the first AES round (if you understand the basic mechanics of AES, you’ll understand why). If you increase scope.adc.samples to capture the full AES computation (scope.adc.trig_count tells you how many samples are needed; again this is target and compiler dependent but on the SAM4S should be around 32K), then you’ll be able to roughly “see” the rounds, and confirm for yourself what I stated.