some rst b ounds on the degree a b ound on the degree of
play

Some rst b ounds on the degree A b ound on the degree of - PowerPoint PPT Presentation

Some rst b ounds on the degree A b ound on the degree of SPN onstrutions Inuene of the inverse p ermutation Bounds on the algeb rai degree of iterated onstrutions Christina Boura DTU Compute June 10,


  1. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Bounds on the algeb rai degree of iterated onstru tions Christina Boura DTU Compute June 10, 2013 1 / 43

  2. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation fun tion F : F n 2 → F n 2 Algeb rai degree of a ve to rial of F 4 ermutation F 2 Example ( ANF of a p ) ( y 0 , y 1 , y 2 , y 3 ) = F ( x 0 , x 1 , x 2 , x 3 ) y 0 = x 0 x 2 + x 1 + x 2 + x 3 = x 0 x 1 x 2 + x 0 x 1 x 3 + x 0 x 2 x 3 + x 1 x 2 + x 0 x 3 + x 2 x 3 + x 0 + x 2 y 1 y 2 = x 0 x 1 x 3 + x 0 x 2 x 3 + x 1 x 2 + x 1 x 3 + x 2 x 3 + x 0 + x 1 + x 3 = x 0 x 1 x 2 + x 1 x 3 + x 0 + x 1 + x 2 + 1 . y 3 is 3 of F The algeb rai degree . 2 / 43

  3. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Iterated p ermutations Most of the symmetri onstru tions (hash fun tions, blo k iphers) a re based on a p ermutation iterated a high numb er of times . Imp o rtant to estimate the algeb rai degree of su h iterated p ermutations. F un tions with a lo w degree a re vulnerable to: • Algeb rai atta ks • Higher-o rder di�erential atta ks and distinguishers • Cub e atta ks 3 / 43

  4. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Higher-o rder derivatives Let F : F n 2 → F m 2 . at a ∈ F n of F : D a ( x ) = F ( x ) ⊕ F ( x + a ) 2 Derivative . of F n any k subspa e V the k 2 De�nition. F o r -dimensional , -th of F to V o rder derivative with resp e t is the fun tion de�ned b y � x ∈ F n D V F ( x ) = D a 1 . . . D a k ( x ) = F ( x + v ) , 2 . fo r every v ∈ V where ( a 1 , . . . , a k ) of V is a basis . , V = � a, b � ( k = 2 Example: ) D V ( x ) = D a D b ( x ) = D a ( F ( x ) ⊕ F ( x + b )) = F ( x ) ⊕ F ( x + a ) ⊕ F ( x + b ) ⊕ F ( x + a + b ) 4 / 43

  5. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Higher-o rder di�erential ryptanalysis Intro du ed b y Knudsen in 1994. Based on the follo wing p rop erties: Let F : F n 2 → F m degree d 2 of . every a ∈ F n 2 Prop osition. F o r w e have D a F ≤ d − 1 . every V ⊂ F n with dim V > d 2 Prop osition. [Lai 94℄ F o r , every x ∈ F n D V ( x ) = 0 , 2 . fo r 5 / 43

  6. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation The KN ipher [Knudsen � Nyb erg 95℄ 6 x i − 1 y i − 1 -round F eistel ipher k i • E : F 32 2 → F 33 2 T S E linea r • T : F 33 2 → F 32 2 linea r • k i : 33 -bit subk ey • S : x �→ x 3 over F 33 2 x i y i of S Algeb rai degree : 2 6 / 43

  7. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation on KN Higher-o rder di�erential atta k [Jak obsen � Knudsen 97℄ x 0 = x y 0 = c y 0 ( x ) = c F k 1 x + F k 1 ( c ) := x + c ′ y 1 ( x ) = F k 2 ( x + c ′ ) + c y 2 ( x ) = F k 2 d = 1 F k 3 ( F k 2 ( x + c ′ ) + c ) + x + c ′ y 3 ( x ) = F k 4 ( F k 3 ( F k 2 ( x + c ′ ) + c ) + x + c ′ ) F k 3 y 4 ( x ) = d = 2 F k 2 ( x + c ′ ) + c + F k 4 d = 4 y 4 G = F k 4 ◦ F k 3 ◦ F k 2 . F k 5 d = 8 F k 6 deg( G ) ≤ 2 3 y 6 x 6 7 / 43

  8. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation If V ⊂ F 32 with dim( V ) = 9 2 , then: all x ∈ F 32 D V y 4 ( x ) = 0 , 2 . fo r By de�nition: � all w ∈ F 32 y 4 ( v + w ) = 0 , 2 . fo r (1) v ∈ V W e an see that: x 6 ( x ) = F k 6 ( y 6 ( x )) + y 4 ( x ) , and b y inverting the terms: y 4 ( x ) = x 6 ( x ) + F k 6 ( y 6 ( x )) . (2) 8 / 43

  9. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Key re overy By ombining equations (1 ) and (2), w e obtain the atta k equation: � � F k 6 ( y 6 ( v + w )) + x 6 ( v + w ) = 0 . v ∈ V v ∈ V ey k 6 The right subk is the one fo r whi h the equation is veri�ed. Complexit y of the atta k: 2 9 • Data Complexit y: plaintexts. 2 33+8 • Time Complexit y: . y 2 5 and 2 9 Distinguisher fo r 4 and 5 rounds with data omplexit resp e tively . 9 / 43

  10. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation SHA-3 [Bertoni � Daemen � P eeters � V anAss he 08℄ Sp onge onstru tion Ke ak - f P ermutation • 1600 a 3 -bit state, seen as -dimensional 5 × 5 × 64 matrix • 24 rounds R • er: 320 Nonlinea r la y pa rallel appli ations a 5 × 5 x χ of S-b o , deg χ − 1 = 3 • deg χ = 2 10 / 43

  11. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Outline Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation 11 / 43

  12. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Outline Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation 12 / 43

  13. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation A trivial b ound from F n into F n Let F and G 2 2 Prop osition: b e a fun tion a from F n into F m 2 2 fun tion . Then deg( G ◦ F ) ≤ deg( G ) deg( F ) . fun tion R degree 7 Example: Round of AES is of . Then deg( R 2 ) = deg( R ◦ R ) ≤ 7 2 = 49 . 13 / 43

  14. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation A b ound based on the W alsh sp e trum [Canteaut � Videau '02℄ of F : F n 2 → F n 2 De�nition (W alsh sp e trum ) � ( − 1) b · F ( x )+ a · x , a, b ∈ F n {F ( ϕ b ◦ F + ϕ α ) = 2 , b � = 0 } . x ∈ F n 2 of F Theo rem: If all the values in the W alsh sp e trum a re divisible y 2 ℓ every G : F n 2 → F n 2 b , then fo r deg( G ◦ F ) ≤ n − ℓ + deg( G ) . 14 / 43

  15. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Appli ation to SHA-3 It an b e omputed that: and χ − 1 y 2 3 • of χ The W alsh sp e tra a re divisible b . re 320 of χ As there a pa rallel appli ations in a round w e have: y 2 3 · 320 = 2 960 and R − 1 • of R The W alsh sp e tra a re divisible b . of R − 7 Bound fo r the degree deg( R − 7 ) = deg( R − 6 ◦ R − 1 ) ≤ 1600 − 960+deg( R − 6 ) ≤ 1369 . 15 / 43

  16. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Appli ation to SHA-3 It an b e omputed that: and χ − 1 y 2 3 • of χ The W alsh sp e tra a re divisible b . re 320 of χ As there a pa rallel appli ations in a round w e have: y 2 3 · 320 = 2 960 and R − 1 • of R The W alsh sp e tra a re divisible b . of R − 7 Bound fo r the degree deg( R − 7 ) = deg( R − 6 ◦ R − 1 ) ≤ 1600 − 960+deg( R − 6 ) ≤ 1369 . deg( R 7 ) ≤ min(1599 , 2187) 15 / 43

  17. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Outline Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation 16 / 43

  18. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation Substitution P ermutation Net w o rks S S S S S S Linear Layer S S S S S S Linear Layer S S S S S S Linear Layer Ho w to estimate the evolution of the degree of su h onstru tions? 17 / 43

  19. Some �rst b ounds on the degree A b ound on the degree of SPN onstru tions In�uen e of the inverse p ermutation x 0 x 1 x 3 x 4 x 5 x 6 x 6 x 7 x 8 x 9 x 10 x 11 x 12 x 13 x 14 x 15 S 1 S 2 S 3 S 4 y 0 y 1 y 2 y 3 y 4 y 5 y 6 y 7 y 8 y 9 y 10 y 11 y 12 y 13 y 14 y 15 After several rounds, all o o rdinates an b e exp ressed as a sum of monomials. in X = { x 0 , . . . , x 15 } Ea h monomial is a p ro du t of va riables . 18 / 43

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend