Welcome!
To use the personalized features of this site, please log in or register.
If you have forgotten your username or password, we can help.
My Menu
Saved Items

Efficient Implementation of Cryptosystems Based on Non-maximal Imaginary Quadratic Orders

Detlef HühnleinContact Information

(6)  secunet Security Networks AG, Mergenthalerallee 77-81, D-65760 Eschborn, Germany
Abstract
In [14] there is proposed an ElGamal-type cryptosystem based on non-maximal imaginary quadratic orders with trapdoor decryption. The trapdoor information is the factorization of the non-fundamental discriminant Δp = Δ1 p 2. The NICE-cryptosystem (New Ideal Coset En-cryption) [24, 12] is an efficient variant thereof, which uses an element $$
\mathfrak{g}^k {\mathbf{ }} \in {\mathbf{  }}Ker(\emptyset _{cl}^{ - 1} ){\mathbf{ }} \subseteq {\mathbf{ }}Cl(\Delta _p )
$$ , where k is random and φ Cl −1 : Clp) → (Δ1) is a map between the class groups of the non-maximal and maximal order, to mask the message in the ElGamal cryptosystem. This mask simply “disappears” during decryption, which essentially consists of computing φ Cl −1 . Thus NICE features quadratic decryption time and hence is very well suited for applications in which a central server has to decrypt a large number of ciphertexts in a short time. In this work we will introduce an efficient batch decryption method for NICE, which allows to speed up the decryption by about 30% for a batch size of 100 messages.
In [17] there is proposed a NICE-Schnorr-type signature scheme. In this scheme one uses the group Ker(φ Cl −1 ) instead of IF p * . Thus instead of modular arithmetic one would need to apply standard ideal arithmetic (multiply and reduce) using algorithms from [5] for example. Because every group operation needs the application of the Extended Euclidean Algorithm the implementation would be very inefficient. Especially the signing process, which would typically be performed on a smartcard with limited computational power would be too slow to allow practical application. In this work we will introduce an entirely new arithmetic for elements in Ker(φ Cl −1 ), which uses the generator and ring-equivalence for exponentiation. Thus the signer essentially performs the exponentiation in ( $$
(\mathcal{O}_{\Delta _1 } /p\mathcal{O}_{\Delta _1 } )*
$$ , which turns out to be about twenty times as fast as conventional ideal arithmetic. Furthermore in [17] it is shown, how one can further speed up this exponentiation by application of the Chinese Remainder Theorem for $$
(\mathcal{O}_{\Delta _1 } /p\mathcal{O}_{\Delta _1 } )*
$$ . With this arithmetic the signature generation is about forty times as fast as with conventional ideal arithmetic and more than twice as fast as in the original Schnorr scheme [26].

Contact Information Detlef Hühnlein
Email: huehnlein@secunet.de
Fulltext Preview (Small, Large)
Image of the first page of the fulltext

References secured to subscribers.



Export this chapter
Export this chapter as RIS | Text
 
Remote Address: 38.107.191.105 • Server: mpweb04
HTTP User Agent: CCBot/1.0 (+http://www.commoncrawl.org/bot.html)