New automatic tool for finding impossible differentials and zero-correlation linear approximations

Cui, Tingting; Chen, Shiyao; Fu, Kai; Wang, Meiqin; Jia, Keting

doi:10.1007/s11432-018-1506-4

Cited by 21 publications

(39 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[21] proposed a model including five inequalities in three input/output binary variables and an extra dummy binary variable that precisely describes the XOR operation. However, considering that all variables in the model are binary‐valued, it was shown in [18] that the following single linear equation completely describes the XOR operation:

a + b + c = 2 d

where d is a dummy binary variable. Since XOR is a linear operation, it does not have any effect on the characteristic probability and hence the objective function.…”

Section: Milp Model For Differential Characteristics In Arx Ciphersmentioning

confidence: 99%

“…Therefore, it has been a focus of cryptographers’ concerns to automise the algorithms of searching the optimal differential characteristics. In this regard, a variety of methods have been proposed and applied such as the methods adopted from the branch and bound Matsui's algorithm [12, 13], the methods based on Boolean satisfiability problems [14–16], and the methods based on mixed‐integer linear programming (MILP) problems [17, 18].…”

Section: Introductionmentioning

confidence: 99%

“…In this study, we focus on the third technique which has been explicitly applied for automated search algorithms for cryptanalysis of symmetric ciphers either SPN or ARX structures [17–25]. MILP is a class of optimisation problems derived from linear programming (LP), which aims to optimise an objective function under a certain set of constraints.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MILP‐based automatic differential search for LEA and HIGHT block ciphers

Bagherzadeh

Ahmadian

2020

IET Information Security

View full text Add to dashboard Cite

a + b + c = 2 d

where d is a dummy binary variable. Since XOR is a linear operation, it does not have any effect on the characteristic probability and hence the objective function.…”

Section: Milp Model For Differential Characteristics In Arx Ciphersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MILP‐based automatic differential search for LEA and HIGHT block ciphers

Bagherzadeh

Ahmadian

2020

IET Information Security

View full text Add to dashboard Cite

“…Cui et al [6] gave the MILP-based impossible differential search model independently. There are many more MILPbased tools proposed recently, such as MILP-based differential/linear search model for ARX ciphers [7], MILP-based conditional cube attacks [8], [9] on Keccak [10], etc.…”

Section: Introductionmentioning

confidence: 99%

“…In the aspect of searching impossible differential, Sasaki et al [5]'s MILP model could not cover the search of related-key/tweakey impossible differential. Cui et al [6] studied a special case of related-key impossible differential of LBlock. However, it is not a general model and could not be applied to other ciphers trivially.…”

Section: Introductionmentioning

confidence: 99%

MILP-Aided Related-Tweak/Key Impossible Differential Attack and its Applications to QARMA, Joltik-BC

Zong

Dong

2019

IEEE Access

View full text Add to dashboard Cite

In this paper, we study the relation of related-tweak/key impossible differentials with singlekey ones. Following a heuristic strategy, we can derive longer related-tweak/key impossible differentials from single-key ones. We implement this strategy with the MILP technique and apply it to search relatedtweak/key impossible differentials of two tweakable block ciphers: QARMA-64 and Joltik-BC-128. For QARMA-64, we find several 7-round related-tweak impossible differential distinguishers and use them to mount a 10-round key recovery attack including the outer whitening key; for Joltik-BC-128, we find two 6-round related-tweakey impossible differential distinguishers and use them attack 9-round and 10-round Joltik-BC-128 respectively.INDEX TERMS Impossible differential attack, Joltik-BC-128, QARMA-64, related-tweakey, tweakable block cipher.

show abstract

Out of Non-linearity: Search Impossible Differentials by the Bitwise Characteristic Matrix

Yang

Shen

Sun

2021

Information Security Practice and Experience

View full text Add to dashboard Cite

In this paper, we propose the M-method which uses the bitwise characteristic matrix to search impossible differentials. M-method exploits not only the linear components but also partial information of non-linear components. According to the principle of miss-in-the-middle, we construct two different types of contradiction to search the impossible differentials with limited time and memory complexity by calculating M r 1 en and M r 2 de which represent r1 rounds encryption and r2 rounds decryption, respectively. Compared with the previous methods, our technique is comprehensible and fast especially for large block size.As a result, we find the 7-round impossible differentials of GIFT-128, the 5-round impossible differentials of PRIDE, and the 4-round impossible differentials of Pyjamask-96. For GIFT-64, PRESENT, RECTANGLE which are well-analyzed by MILP-method or SAT-method, we construct new impossible differentials. Moreover, the efficiency of our method will not be influenced by the block size, which makes us find the new 5-round impossible differentials of the 320-bit permutation of ASCON.

show abstract

New automatic tool for finding impossible differentials and zero-correlation linear approximations

Cited by 21 publications

References 6 publications

MILP‐based automatic differential search for LEA and HIGHT block ciphers

MILP‐based automatic differential search for LEA and HIGHT block ciphers

MILP-Aided Related-Tweak/Key Impossible Differential Attack and its Applications to QARMA, Joltik-BC

Out of Non-linearity: Search Impossible Differentials by the Bitwise Characteristic Matrix

Contact Info

Product

Resources

About