To address the absence of a systematic method for effectively guiding practical attacks on small CRT-RSA, a practical attack method based on parameter optimization, fitting and binary search is proposed. Initially, the parameters and in the small attack are optimized. Subsequently, guided by the optimized values of and , along with experimental statistics on the length of the reduced lattice basis, an estimation method for an attackable upper bound is proposed through parameter fitting, which aligns with experimental results. Guided by the method, the attackable upper bound of small given the parameters is explored. Furthermore, the most significant bits (MSBs) guess strategy for the prime factor of the modulus, which is proposed in previous explorations of practical attacks on small RSA decryption exponents, is introduced. Based on the “multivalued phenomenon” observed in experiments, a binary search attack method suitable for parallel implementation is proposed, effectively enhancing the upper bound of practical attacks on small under the given parameters. Additionally, the lattice proposed by Takayasu et al. is optimized, and the performance of helpful polynomials in practical attacks is explored. The practical attack exploration presented here can offer inspiration and assistance for the future research on small attacks against CRT-RSA.
RIVESTR L, SHAMIRA, ADLEMANL. A method for obtaining digital signatures and public-key cryptosystems[J]. Communications of the ACM, 1978,21(2):120-126.
[2]
WIENERM J. Cryptanalysis of short RSA secret exponents[C]∥Proceedings of the Advances in Cryptology- EUROCRYPT ’89. Berlin, Germany: Springer, 1990:372.
[3]
STINSOND R. Cryptography:theory and practice[M]. 2nd ed. Florida,USA: CRC Press, 2002.
[4]
MAY A. Computing the RSA secret key is deterministic polynomial time equivalent to factoring[C]∥Proceedings of the Advances in Cryptology-CRYPTO 2004. Berlin, Germany: Springer, 2004:213-219.
[5]
CORONJ S, MAY A. Deterministic polynomial-time equivalence of computing the RSA secret key and factoring[J]. Journal of Cryptology, 2007,20(1):39-50.
[6]
BONEHD, DURFEEG. Cryptanalysis of RSA with private key d less than N0.292[C]∥Proceedings of the Advances in Cryptology-EUROCRYPT’99. Berlin, Germany: Springer, 1999:1-11.
[7]
DONC. Finding a small root of a univariate modular equation[C]∥Proceedings of the Advances in Cryptology-EUROCRYPT’96. Berlin, Germany: Springer, 1996:155-165.
GALBRAITHS D, HENEGHANC, MCKEEJ F. Tunable balancing of RSA[M]∥Information security and privacy. Berlin, Germany: Springer, 2005:280-292.
[10]
BLEICHENBACHERD, MAY A. New attacks on RSA with small secret CRT-exponents[C]∥Proceedings of the Public Key Cryptography-PKC 2006. Berlin, Germany: Springer, 2006:1-13.
JOCHEMSZE, MAY A. A polynomial time attack on RSA with private CRT-exponents smaller than N0.073[C]∥Proceedings of the Advances in Cryptology-CRYPTO 2007. Berlin, Germany: Springer, 2007:395-411.
[13]
TAKAYASUA, LUY, PENGL Q. Small CRT-exponent RSA revisited[M]∥Advances in Cryptology-EUROCRYPT 2017. Cham, Switzerland: Springer, 2017:130-159.
[14]
TAKAYASUA, LUY, PENGL Q. Small CRT-exponent RSA revisited[J]. Journal of Cryptology, 2019,32(4):1337-1382.
[15]
PENGL Q, TAKAYASUA. Generalized cryptanalysis of small CRT-exponent RSA[J]. Theoretical Computer Science, 2019,795:432-458.
[16]
MAY A, NOWAKOWSKIJ, SARKARS. Partial key exposure attack on short secret exponent CRT-RSA[M]∥Advances in Cryptology-ASIACRYPT 2021. Cham, Switzerland: Springer, 2021: 99-129.
[17]
ZHOUY Y, VAN DEPOL J, YUY, et al. A third is all you need: extended partial key exposure attack on CRT-RSA with additive exponent blinding[M]∥Advances in cryptology-ASIACRYPT 2022. Cham, Switzerland: Springer, 2022:508-536.
[18]
MAY A, NOWAKOWSKIJ, SARKARS. Approximate divisor multiples-factoring with only a third of the secret CRT-exponents[M]∥Advances in Cryptology-EUROCRYPT 2022. Cham, Switzerland: Springer, 2022:147-167.
[19]
MAY A. Cryptanalysis of unbalanced RSA with small CRT-exponent[C]∥Proceedings of the Advances in Cryptology-CRYPTO 2002. Berlin, Germany: Springer, 2002:242-256.
HERRMANNM, MAY A. Maximizing small root bounds by linearization and applications to small secret exponent RSA[C]∥/Proceedings of the Public Key Cryptography-PKC 2010. Berlin, Germany: Springer, 2010:53-69.
[22]
NGUYENP Q, STEHLÉD. LLL on the average[C]∥Proceedings of the Algorithmic Number Theory. Berlin, Germany: Springer,2006:238-256.
[23]
HOWGRAVE-GRAHAMN. Finding small roots of univariate modular equations revisited[C]∥Proceedings of the Crytography and Coding. Berlin, Germany: Springer, 1997:131-142.
[24]
TAKAYASUA, KUNIHIRON. Better lattice constructions for solving multivariate linear equations modulo unknown divisors[C]∥Proceedings of the Information Security and Privacy. Berlin, Germany: Springer, 2013:118-135.
[25]
MAY A. Using LLL-reduction for solving RSA and factorization problems[M]∥The LLL algorithm. Berlin, Germany: Springer, 2009:315-348.
[26]
LIQ, ZHENGQ X, QIW F. Practical attacks on small private exponent RSA: new records and new insights[J]. Designs, Codes and Cryptography, 2023,91(12):4107-4142.