-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathAES_mia_sca_16x.py
112 lines (91 loc) · 5.41 KB
/
AES_mia_sca_16x.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
# Mutual Information Analysis LIKE* practice on AES S-box based on Hamming weights:
# multiple bytes implementation, where at hacking a specific input byte, all other
# byte substitutions represent algorithmic noise
# (*) As I interpret Shannon entropy based MIA, it is something like finding out if
# the probability distribution ie. the frequencies of output samples relate to choosen
# input data as expected. I left out logarithm for example, because it is a monothonic
# function, summing up logarithms has to have the same correlation results as just plain
# numbers. I rather gather the corresponding leakage to every plain text value. Now since
# all leakages are between 0 to 8 per S-box output, while there are 256 possible plain
# text options, the matching is between which plain text causes what amount of leakage
# with a certain key assumed.
# As for the results, roughly the same amount of samples are needed for a proper key
# recovery, but the calculation got a lot faster compared to CPA SCA.
import random
import time
s_box = [
0x63, 0x7C, 0x77, 0x7B, 0xF2, 0x6B, 0x6F, 0xC5, 0x30, 0x01, 0x67, 0x2B, 0xFE, 0xD7, 0xAB, 0x76,
0xCA, 0x82, 0xC9, 0x7D, 0xFA, 0x59, 0x47, 0xF0, 0xAD, 0xD4, 0xA2, 0xAF, 0x9C, 0xA4, 0x72, 0xC0,
0xB7, 0xFD, 0x93, 0x26, 0x36, 0x3F, 0xF7, 0xCC, 0x34, 0xA5, 0xE5, 0xF1, 0x71, 0xD8, 0x31, 0x15,
0x04, 0xC7, 0x23, 0xC3, 0x18, 0x96, 0x05, 0x9A, 0x07, 0x12, 0x80, 0xE2, 0xEB, 0x27, 0xB2, 0x75,
0x09, 0x83, 0x2C, 0x1A, 0x1B, 0x6E, 0x5A, 0xA0, 0x52, 0x3B, 0xD6, 0xB3, 0x29, 0xE3, 0x2F, 0x84,
0x53, 0xD1, 0x00, 0xED, 0x20, 0xFC, 0xB1, 0x5B, 0x6A, 0xCB, 0xBE, 0x39, 0x4A, 0x4C, 0x58, 0xCF,
0xD0, 0xEF, 0xAA, 0xFB, 0x43, 0x4D, 0x33, 0x85, 0x45, 0xF9, 0x02, 0x7F, 0x50, 0x3C, 0x9F, 0xA8,
0x51, 0xA3, 0x40, 0x8F, 0x92, 0x9D, 0x38, 0xF5, 0xBC, 0xB6, 0xDA, 0x21, 0x10, 0xFF, 0xF3, 0xD2,
0xCD, 0x0C, 0x13, 0xEC, 0x5F, 0x97, 0x44, 0x17, 0xC4, 0xA7, 0x7E, 0x3D, 0x64, 0x5D, 0x19, 0x73,
0x60, 0x81, 0x4F, 0xDC, 0x22, 0x2A, 0x90, 0x88, 0x46, 0xEE, 0xB8, 0x14, 0xDE, 0x5E, 0x0B, 0xDB,
0xE0, 0x32, 0x3A, 0x0A, 0x49, 0x06, 0x24, 0x5C, 0xC2, 0xD3, 0xAC, 0x62, 0x91, 0x95, 0xE4, 0x79,
0xE7, 0xC8, 0x37, 0x6D, 0x8D, 0xD5, 0x4E, 0xA9, 0x6C, 0x56, 0xF4, 0xEA, 0x65, 0x7A, 0xAE, 0x08,
0xBA, 0x78, 0x25, 0x2E, 0x1C, 0xA6, 0xB4, 0xC6, 0xE8, 0xDD, 0x74, 0x1F, 0x4B, 0xBD, 0x8B, 0x8A,
0x70, 0x3E, 0xB5, 0x66, 0x48, 0x03, 0xF6, 0x0E, 0x61, 0x35, 0x57, 0xB9, 0x86, 0xC1, 0x1D, 0x9E,
0xE1, 0xF8, 0x98, 0x11, 0x69, 0xD9, 0x8E, 0x94, 0x9B, 0x1E, 0x87, 0xE9, 0xCE, 0x55, 0x28, 0xDF,
0x8C, 0xA1, 0x89, 0x0D, 0xBF, 0xE6, 0x42, 0x68, 0x41, 0x99, 0x2D, 0x0F, 0xB0, 0x54, 0xBB, 0x16]
# Hamming weight of x
def count1s(x):
y = 0
while(x>0):
y+=(x&1)
x>>=1
return y
# PREPARATION
# -----------
print("Start hypothesis:", time.time())
# hypothesises of hamming weights for keys
hypothesises = [[None for j in range(256)] for i in range(256)] #1st index: key; 2nd index: plain texts
for plain in range(256):
for key in range(256):
hypothesises[key][plain] = count1s(s_box[plain ^ key])
# SIMULATE MEASUREMENTS
# ---------------------
print("Start simulation:", time.time())
# number of choosen plain texts
number_of_traces = 70000 # 30000 already gives some poor results
# number of parallel S-box computations
S_box_count = 16
# picking a set of random keys
key_bytes = [random.randint(0,255) for i in range(S_box_count)]
# picking random input data sets
choosen_plain_texts = [[random.randint(0,255) for j in range(number_of_traces)] for i in range(S_box_count)]
# calculating leakages
leakages = [0 for i in range(number_of_traces)]
for tr in range(number_of_traces):
for sb in range(S_box_count):
leakages[tr] += count1s(s_box[key_bytes[sb] ^ choosen_plain_texts[sb][tr]])
print('Real key bytes:', key_bytes)
# HACK IT
# -------
print("Start hacking:", time.time())
#calculating something like mutual information
most_likely_keys = [None for i in range(S_box_count)]
# setting up probability masses of leakages with respect to choosen plain texts: each plain text value should trigger a certain
# amount of leakage, so summing up leakages at given plain text values gives a probability distribution characteristic to the key
# loop through all keys
for key_idx in range(S_box_count):
# assigning leakages to their respective(*line 104) choosen plain text byte values and simultaneously summing up the corresponding leakage values
histogram_like_stuff = [0 for i in range(256)] # neither histogram, nor a probability mass, as described in line 9
for tr in range(number_of_traces):
histogram_like_stuff[choosen_plain_texts[key_idx][tr]] += leakages[tr]
# find best matching between of histogram and hypothesises
current_likely_key = 0
highest_correlation = 0
for assumed_key in range(256):
correlation = 0
for h2h_idx in range(256):
# (*line 93) it is vital, that all S-boxes have different and known choosen plain text series, which result in different "histograms"
correlation += hypothesises[assumed_key][h2h_idx] * histogram_like_stuff[h2h_idx]
if correlation > highest_correlation:
highest_correlation = correlation
current_likely_key = assumed_key
most_likely_keys[key_idx] = current_likely_key
print('Most likely keys:', most_likely_keys)
print("Done:", time.time())