高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

理想格上格基的快速三角化算法研究

张洋 刘仁章 林东岱

张洋, 刘仁章, 林东岱. 理想格上格基的快速三角化算法研究[J]. 电子与信息学报, 2020, 42(1): 98-104. doi: 10.11999/JEIT190725
引用本文: 张洋, 刘仁章, 林东岱. 理想格上格基的快速三角化算法研究[J]. 电子与信息学报, 2020, 42(1): 98-104. doi: 10.11999/JEIT190725
Yang ZHANG, Renzhang LIU, Dongdai LIN. Fast Triangularization of Ideal Latttice Basis[J]. Journal of Electronics & Information Technology, 2020, 42(1): 98-104. doi: 10.11999/JEIT190725
Citation: Yang ZHANG, Renzhang LIU, Dongdai LIN. Fast Triangularization of Ideal Latttice Basis[J]. Journal of Electronics & Information Technology, 2020, 42(1): 98-104. doi: 10.11999/JEIT190725

理想格上格基的快速三角化算法研究

doi: 10.11999/JEIT190725
详细信息
    作者简介:

    张洋:男,1991年生,博士生,研究方向为基于理想格算法的密码算法分析

    刘仁章:男,1989年生,博士,研究方向为格算法及格密码算法分析

    林东岱:男,1964年生,研究员,研究方向为密码学与安全协议、网络与系统安全、分布式密码计算

    通讯作者:

    张洋 zhangyang9091@iie.ac.cn

  • 中图分类号: TP309.7; O157.4

Fast Triangularization of Ideal Latttice Basis

  • 摘要:

    为了提高理想格上格基的三角化算法的效率,该文通过研究理想格上的多项式结构提出了一个理想格上格基的快速三角化算法,其时间复杂度为O(n3log2B),其中n是格基的维数,B是格基的无穷范数。基于该算法,可以得到一个计算理想格上格基Smith标准型的确定算法,且其时间复杂度也比现有的算法要快。更进一步,对于密码学中经常所使用的一类特殊的理想格,可以用更快的算法将三角化矩阵转化为格基的Hermite标准型。

  • 表  1  本原格向量序列

     输入:$ {\mathbb{Z}}\left[ {{x}} \right]$中$ f\left( x \right)$和$ g\left( x \right)$,次数分别为$ n$和$ m$,且$ n>m$;
        (1) 利用扩展欧几里得算法计算$ {\mathbb{Q}}[x]$上$ {r_i}^{'}\left( x \right)$, $ {s_i}^{'}\left( x \right)$,      $ {t_i}^{'}\left( x \right)$,使得$ {r_i}^{'}\left( x \right) = {r_{i - 2}}^{'}\left( x \right) + {q_i}^{'}\left( x \right){r_{i - 1}}^{'}\left( x \right)$      和${r_i}^{'}\left( x \right) = {s_i}^{'}\left( x \right)f\left( x \right) + $$ {t_i}^{'}\left( x \right)g\left( x \right)$成立,这里      $ i = 1,2, ··· ,l$;
        (2) 计算每一个$ {t_i}^{'}\left( x \right)$系数分母的最小公倍数$ {C_i}$,       $ i = 1,2, ··· ,l$;
        (3) 令$ {r_i}\left( x \right) = {r_i}^{'}\left( x \right) \cdot {C_i}$为余式序列中第$ i$个余式,      $ i = 1,2, ··· ,l$;
     输出:$ {r_1}\left( x \right)$, $ {r_2}\left( x \right)$, ···, $ {r_l}\left( x \right)$
    下载: 导出CSV

    表  2  理想格上格基的三角化

     输入:本原格向量序列,$ {r_0}\left( x \right)$, $ {r_1}\left( x \right)$, ···, $ {r_l}\left( x \right)$(向量形式为$ {{{r}}_{\bf 0}}$,
        $ {{{r}}_{\bf 1}}$, ···, $ {{{r}}_{ l}}$)
        (1) 令$ {{T}} \leftarrow {0^{n \times n}}$
        (2) 如果$ k \in {I_l},{{ T}_k}\left( x \right) = {r_l}\left( x \right){x^{n - k}},i \leftarrow l - 1$
        (3) 如果$ k \in {I_i}$,
        (a) 计算$ \phi $和ψ使得
          $ \phi {\rm{lc}}\left( {{r_i}} \right) + \psi {\rm{lc}}\left( {{{{T}}_{n - {n_{i + 1}}}}} \right) = {\rm{gcd}}\left( {\rm{lc}}\left( {{{{T}}_{n - {n_{i + 1}}}}} \right),\right.$
          $\left.{\rm{lc}}\left( {{r_i}} \right) \right) $
        (b) 令$ {{{T}}_{n - {n_i}}}\left( {{x}} \right) = \phi {r_i}\left( x \right) + \psi {{{T}}_{n - {n_{i + 1}}}}\left( x \right){x^{{\delta _i}}}$
        (c) 如果$ {\rm{lc}}\left( {{{{T}}_{n - {n_i}}}} \right) = 1$,则令
          $ {{{T}}_j}\left( x \right) = {{{T}}_{n - {n_i}}}\left( x \right){x^{n - {n_i} - j}}$,
          $ j = 1,2, ··· ,n - {n_i}$,并结束循环
        (d) 否则$ {{{T}}_k}\left( x \right) = {{{T}}_{n - {n_i}}}\left( x \right){x^{n - {n_i} - k}}$, $ i \leftarrow i - 1$
        (e) 如果$ i>0$,到(3)开始循环,否则结束循环
     输出:$ {{T}}$
    下载: 导出CSV
  • FRUMKIN M A. Complexity questions in number theory[J]. Journal of Soviet Mathematics, 1985, 29(4): 1502–1517. doi: 10.1007/bf02104748
    HUNG M S and ROM W O. An application of the Hermite normal form in integer programming[J]. Linear Algebra and its Applications, 1990, 140: 163–179. doi: 10.1016/0024-3795(90)90228-5
    HAFNER J L and MCCURLEY K S. A rigorous subexponential algorithm for computation of class groups[J]. Journal of the American Mathematical Society, 1989, 2(4): 837–850. doi: 10.1090/S0894-0347-1989-1002631-0
    HARTLEY B and HAWKES T O. Rings, Modules and Linear Algebra[M]. London: Chapman and Hall, 1970: 73.
    MICCIANCIO D. Improving lattice based cryptosystems using the Hermite normal form[C]. International Conference on Cryptography and Lattices, Providence, 2001: 126–145. doi: 10.1007/3-540-44670-2_1.
    LYUBASHEVSKY V and PREST T. Quadratic time, linear space algorithms for Gram-Schmidt orthogonalization and Gaussian sampling in structured lattices[C]. The 34th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Sofia, 2015: 789–815. doi: 10.1007/978-3-662-46800-5_30.
    HAFNER J L and MCCURLEY K S. Asymptotically fast triangulation of matrices over rings[C]. The 1st Annual ACM-SIAM Symposium on Discrete Algorithm, San Francisco, 1990: 197–200.
    LE GALL F. Powers of tensors and fast matrix multiplication[C]. The 39th International Symposium on Symbolic and Algebraic Computation, Kobe, 2014: 296–303. doi: 10.1145/2608628.2608664.
    STORJOHANN A and LABAHN G. Asymptotically fast computation of Hermite normal forms of integer matrices[C]. 1996 International Symposium on Symbolic and Algebraic Computation, Zurich, 1996: 259–266.
    DING Jintai and LINDNER R. Identifying ideal lattices[EB/OL]. http://eprint.iacr.org/2007/322, 2007.
    ZHANG Yang, LIU Renzhang, and LIN Dongdai. Improved key generation algorithm for Gentry’s fully homomorphic encryption scheme[C]. The 20th International Conference on Information Security and Cryptology, Seoul, 2018: 93–111. doi: 10.1007/978-3-319-78556-1_6.
    VON ZUR GATHEN J and GARHARD J. Modern Computer Algebra[M]. 3rd ed. Cambridge: Cambridge University Press, 2013: 313–332.
    刘仁章. 格算法及其密码学应用[D]. [博士论文], 中国科学院大学数学与系统科学研究院, 2016.
  • 加载中
表(2)
计量
  • 文章访问数:  3123
  • HTML全文浏览量:  987
  • PDF下载量:  74
  • 被引次数: 0
出版历程
  • 收稿日期:  2019-09-19
  • 修回日期:  2019-11-15
  • 网络出版日期:  2020-01-01
  • 刊出日期:  2020-01-21

目录

    /

    返回文章
    返回