Search

SM2 key exchange protocol is a part of the SM2 public key cryptographic algorithm based on elliptic curves which has been issued by Chinese State Cryptography Administration since 2010. Under the guide of Chinese government, SM2 has been widely used in Chinese commercial applications. This paper gives the first partially known information attack on SM2 key exchange protocol. Our attack is based on a technique modified from the hidden number problem (HNP) which was introduced originally to study the bit security of Diffie-Hellman and related schemes. We present a polynomial-time algorithm which could recover the users secret key when given about half least significant bits of the two unknown intermediate values in each congruence over about 30 to 40 instances. Compared with the standard HNP, our approach deals with congruence involved two independent unknown variables and each of them possesses the same size as the secret key. Moreover, our results almost coincide with the previous best result among the same field considering the extreme case in which one variant is completely revealed.

As a feasible model for signing quantum messages, some cryptanalysis and improvement of arbitrated quantum signature (AQS) have received a great deal of attentions in recent years. However, in this paper we find the previous improvement is not suitable implemented in some typical AQS protocols in the sense that the receiver, Bob, can forge a valid signature under known message attack. We describe the forgery strategy and present some corresponding improved strategies to stand against the forgery attack by modifying the encryption algorithm, an important part of AQS. These works preserve the merits of AQS and lead some potential improvements of the security in quantum signature or other cryptography problems.

With the consistent propulsion of the application software, the protection of software itself becomes a serious issue. A novel algorithm and an approach to behavior self-check for software tamper-proofing is proposed in this paper. To represent the principle of tamper-proofing, a behavior model is used to specify software behavior characteristic while the behavior fingerprint is computed by the secure Hash function. Combined with the program flow controlling and self-decryption technology, behavior self-checking technique can monitor instruction, capture behavior and calculate behavior fingerprint to ensure anti-attack capability of software. Analysis results show that the algorithm is capable of resisting static and dynamic attacks strongly.