The Secret Battle of Encryption Algorithms

Transcript

1. The Secret Battle of Encryption Algorithms by @amandasopkin
4. Disclaimer @amandasopkin
5. Where are we headed? @amandasopkin
6. Cryptography (bad): “The art of writing or solving codes.” @amandasopkin
7. Cryptography (better): “method of storing and transmitting data in a particular form so that only those for whom it is intended can read and process it.” -IEEE
9. @amandasopkin Bread and Butter of encryption
10. Encryption: encoding using a key and an encryption algorithm @amandasopkin
11. Key derivation: process for creating suitable keys for use in encryption @amandasopkin
12. Plain text: input in its natural form Can be stream of bits, text file, bitmap, etc. @amandasopkin
13. Cipher text: input that has been encrypted @amandasopkin
14. Basic process:
15. Ciphers: • Method for disguising text @amandasopkin
16. History @amandasopkin
17. 1900 BC: Cryptographic hieroglyphics @amandasopkin
19. 100 BC: Julius Caeser @amandasopkin
20. @amandasopkin
21. Documentation of Caesar Cipher:
22. Did Caesar design more complicated systems as well?
23. How effective was the Caesar Cipher?
24. How effective was the Caesar Cipher?
25. Effectiveness of Caesar Cipher: Many of Caesar’s enemies were illiterate Others assumed that encoded letters were in another language
26. 9th century Al Kindi (mathematician): Earliest surviving frequency analysis records
27. Letter frequencies Can be used to break a cipher
28. Letter frequencies
29. Letter frequencies Frequency of each letter in cipher Sorted from most common to least
30. Letter frequencies Frequencies of letters in cipher
31. Letter frequencies Standard english letter frequencies
32. Letter frequencies Most frequent letters in cipher are S and O
33. Substitute E and T for S and O
34. Spot instances of “tle”
35. Most common letter is now G, which must be a, i, or o
36. Substitute e and t for s and o and o for g
37. Spot oFe and theF and then Lheet
38. Spot sODVe and OK
39. Sub l for D, v for V, R for K
40. Spot enoQRh
41. Spot EesMge and Nount
42. Few more steps...
43. Solved!
44. 1500s: Vignere’s system @amandasopkin
46. here is a super secure message sups up s upsup supsup supsups Key = “sup”
48. here is a super secure message sups up s upsup supsup supsups z... .. . ..... ...... ....... Key = “sup”
49. How effective was the Vignere system?
50. How effective was the Vignere system?
51. “le chiffre indéchiffrable” until 1863 (but broken as early as the 16th century)
52. How was it broken?
53. The Kasiski examination: takes advantage of the fact that repeated words are, by chance, sometimes encrypted using the same key letters, leading to repeated groups in the ciphertext.
54. Key: ABCDABCDABCDABCDABCDABCDABCD Plaintext: CRYPTOISSHORTFORCRYPTOGRAPHY Ciphertext: CSASTPKVSIQUTGQUCSASTPIUAQJB The Kasiski examination:
55. Key: ABCDABCDABCDABCDABCDABCDABCD Plaintext: CRYPTOISSHORTFORCRYPTOGRAPHY Ciphertext: CSASTPKVSIQUTGQUCSASTPIUAQJB The Kasiski examination: Using the distance between repeated subsequences, the length of the key can be found.
56. Key: ABCDABCDABCDABCDABCDABCDABCD Plaintext: CRYPTOISSHORTFORCRYPTOGRAPHY Ciphertext: CSASTPKVSIQUTGQUCSASTPIUAQJB The Kasiski examination: Once length is known, if a key is N letters long then every Nth letter must have been encoded using same letter of the text
57. Vignere System and The Civil War Confederate soldiers had messages frequently cracked because they relied on the phrases “Complete Victory,” “Manchester Bluff” and “Come Retribution.”
58. Pro Tip: Don’t Use Easy to Guess Phrases!
59. Kerchoff’s Principle: secrecy of key @amandasopkin
60. Let’s get electric! @amandasopkin
61. 1800s: Hebern’s system @amandasopkin
62. 1800s: Hebern’s system
63. 1800s: Hebern’s system Defines a monoalphabet Rotates after each input is processed
64. William Friedman Hebern System vs.
65. Breaking Hebern’s system
66. Breaking Hebern’s system With each rotor, one step = one keypress Fastest rotor always at either end of rotor series Statistical method called kappa test applied
67. William Friedman: 1 vs. Hebern System: 0
68. How effective was Hebern’s system?
69. How effective was Hebern’s system?
70. WW1/WW2: Engima Machine @amandasopkin
72. Breakthroughs by Polish Mathematicians WW1: Engima Machine
73. Marian Rejewski
74. Marian Rejewski vs. Enigma Machine
75. Key Insights into Enigma 1. Single initial 6 letter setting for all messages each day 2. Chosen message key repeated in this initial setting
76. Indicator or Grundstellung = initial rotor setting
77. Indicator or Grundstellung = initial rotor setting Initial setting RAO 3 letter message key IHL Resulting indicator from setting rotors to RAO and encoding IHL twice DQYQQT
78. Indicator: DQYQQT
79. Indicator: DQYQQT D and Q represent the same letter
80. By collecting enough messages enciphered with same indicator, a table could be produced:
81. Path from first letter to fourth then from that letter to its fourth and so on leads to cycle group
82. Path from first letter to fourth then from that letter to its fourth and so on leads to cycle group
83. Path from first letter to fourth then from that letter to its fourth and so on leads to cycle group
84. Path from first letter to fourth then from that letter to its fourth and so on leads to cycle group
85. Cycle group starting at A 9 (A, N, P, L, X, M, J, F, I, A) Cycle group starting at B 3 (B, S, Z, B) Cycle group starting at C 9 (C, Y, D, Q, W, V, O, U, G, C) Cycle group starting at E 3 (E, T, K, E) Cycle group starting at H 1 (H, H) Cycle group starting at R 1 (R, R)
86. Cycle group starting at A (A, N, P, L, X, M, J, F, I, A) Cycle group starting at B (B, S, Z, B) Cycle group starting at C (C, Y, D, Q, W, V, O, U, G, C) Cycle group starting at E (E, T, K, E) Cycle group starting at H (H, H) Cycle group starting at R (R, R) 1* 3 * 9 = 27 possibilities for ciphers at 1 and 4
87. Another weak link = Lazy cipher clerks! Key Insights into Enigma
88. Lazy cipher clerks often used same starting position i.e. “AAA” Key Insights into Enigma
89. Solve for indicator/day key => Factor out board permutation => Commercial Enigma wiring => Rightmost rotor wiring Given a day’s traffic...
90. Solve for indicator/day key => Factor out board permutation => Commercial Enigma wiring => Rightmost rotor wiring Given a day’s traffic...
92. Given a day’s traffic... Solve for indicator/day key => Factor out board permutation => Commercial Enigma wiring => Rightmost rotor wiring
93. Marian Rejewski: 1 vs. Enigma Machine: 0
94. Alan Turing and British efforts to crack the “Engima”
95. Using findings of Rejewski and others:
96. One Bombe = 36 Enigmas!
97. One Bombe = 36 Enigmas! 97,000 parts 12 miles of wiring £4,000,000 2,000 pounds 7 feet wide
98. Bombe provided several possible answers, Codebreakers narrowed it down
99. At peak, 200+ Bombe machines cracking 3,000 messages each day!
100. Shortened the war by an estimated 2 years!
101. How effective was The Engima machine?
102. How effective was The Engima machine?
103. Pro Tip: Establish a process and eliminate possible human error!
104. Let’s get some standards!
105. 1973: IBM’s Luther @amandasopkin
106. Note on block cipher encryption @amandasopkin
108. DES @amandasopkin
109. DES
111. DES Criticism @amandasopkin - Shortened key length (56 bits) - S-box structure
112. DES Criticism “NSA worked closely with IBM to strengthen the algorithm against all except brute-force attacks and to strengthen substitution tables, called S-boxes. Conversely, NSA tried to convince IBM to reduce the length of the key from 64 to 48 bits. Ultimately they compromised on a 56-bit key.” -American Cryptology During the Cold War
113. DES Criticism "We sent the S-boxes off to Washington. They came back and were all different.” -Alan Konheim (one of the designers of DES)
114. 20 years later...
115. "It took the academic community two decades to figure out that the NSA 'tweaks' actually improved the security of DES." -Bruce Schneier
116. How effective was DES?
117. How effective was DES?
118. 2000: Advanced Encryption Standard (Rijndael) @amandasopkin
120. AES = 128 bits! @amandasopkin
121. Attacks on AES @amandasopkin
122. Attacks on AES => so far not practical @amandasopkin
123. Attacks on AES => so far not practical @amandasopkin
124. Attacks on [insert algorithm with x > 256 bit key strength here] => so far not practical @amandasopkin
125. How effective is AES?
126. How effective is AES?
127. Crypto Wars @amandasopkin
129. Key size restrictions @amandasopkin
130. 1991: Sen. Biden introduced a bill requiring providers of electronic communication to provide voice, data, and other content to the government when authorized by law
131. 1991: PGP (Pretty Good Privacy) @amandasopkin
132. PGP: relies on math that is difficult to reverse @amandasopkin
133. PGP = 3 keys 1. Public key 2. Private key 3. Encrypted key that gets sent
135. 1. You generate a random key. 2. You use that key to encrypt your data. 3. I send you my public key. 4. My public key is used to encrypt your random key. 5. You send both the encrypted data and the encrypted random key to me. 6. I use my private key to decrypt your random key. 7. I use your random key to decrypt the data.
136. Phil Zimmerman/PGP vs. Criminal Investigation
137. Published by MIT press to allow export under 1st amendment
138. Phil Zimmerman/PGP: 1 vs. Criminal Investigation: 0
139. The end of crypto wars @amandasopkin
140. Randomness & Encryption @amandasopkin
141. 4oio342ip4o24p32o
142. 4fdslf95454
145. DUAL_EC_DRBG Controversy • 2004: Dual EC PRNG introduced
146. • 08/2007: Shumow and Ferguson present Dual_EC_DRBG flaw at cryptography conference DUAL_EC_DRBG Controversy
147. • 11/2007: Schneier bases article in Wired on their findings DUAL_EC_DRBG Controversy
148. “...would allow NSA to determine the state of the random number generator, and thereby eventually be able to read all data sent over the SSL connection.” DUAL_EC_DRBG Controversy
150. • 09/2013: One of the purposes of Bullrun is described as being "to covertly introduce weaknesses into the encryption standards followed by hardware and software developers around the world." DUAL_EC_DRBG Controversy
151. • NIST recommends removal of the algorithm as a standard DUAL_EC_DRBG Controversy
152. • 2004: Dual EC PRNG introduced • 08/2007: Shumow and Ferguson present Dual_EC_DRBG flaw at cryptography conference • 11/2007: Schneier bases article in Wired on their findings DUAL_EC_DRBG Controversy
153. • 09/2013: One of the purposes of Bullrun is described as being "to covertly introduce weaknesses into the encryption standards followed by hardware and software developers around the world." • 12/2013: Presidential advisory examines encryption standards • 2014: Standard is removed DUAL_EC_DRBG Controversy
154. Years until standard removed... 10!
155. Who did this impact? Microsoft, Google, Apple, McAfee, Docker, IBM, Oracle, Cisco, VMWare, Juniper, HP, Red Hat, Samsung, Toshiba, DELL, Ruckus, F5 Networks, Lenovo, Nokia, the RSA BSAFE libraries for Java and C++ and more....
156. Pro Tip: Don’t assume standardized = good
157. Modern Encryption... Where are we now?
158. Modern Encryption...
159. Modern Encryption... ?
160. Pro Tip: Often good security is not flashy
161. Common breach causes Not encrypting all the things Using cloud storage without pre-encrypting Using a poor random number generator
162. Cloud encryption
163. Let’s wrap up... @amandasopkin
164. Security != obscurity Process is ESSENTIAL Trust no one (kidding) Historical Encryption Lessons @amandasopkin
165. Something to say? Amanda Sopkin @amandasopkin
166. Thank you! @amandasopkin
167. THE END @amandasopkin
168. Sources: • Icons taken from flaticon.com • https://crypto.stackexchange.com/questions/51232/using- 32-hexadecimal-digits-vs-ascii-equivalent-16-character- password • https://dev.to/walker/pseudo-random-numbers-in-python-f rom-arithmetic-to-probability-distributions • Wired Magazine • The Washington Post • NYT • http://home.bt.com/tech-gadgets/cracking-the-enigma-cod e-how-turings-bombe-turned-the-tide-of-wwii-11363990654 704 • Geeks for Geeks • Crypto Corner