Low space complexity CRT-based bit-parallel GF (2n) polynomial basis multipliers for irreducible trinomials

被引:1
|
作者
Zhang, Jiajun [1 ,2 ]
Fan, Haining [2 ,3 ]
机构
[1] Tsinghua Univ, Sch Software, Beijing, Peoples R China
[2] Tsinghua Univ, TNLIST, Beijing, Peoples R China
[3] Tsinghua Univ, Dept Comp Sci & Technol, Beijing, Peoples R China
关键词
Finite field; Multiplication; Polynomial basis; The Chinese Remainder Theorem; Irreducible polynomial; MATRIX-VECTOR PRODUCT; FIELDS;
D O I
10.1016/j.vlsi.2017.02.008
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents new space complexity records for the fastest parallel GF (2(n)) multipliers for about 22% values of n such that a degree-n irreducible trinomial f = u(n) + u(k) + 1 exists over GF (2). By selecting the largest possible value of k epsilon (n/2, 2n/3], we further reduce the space complexities of the Chinese remainder theorem (CRT)-based hybrid polynomial basis multipliers. Our experimental results show that among the 539 values of n epsilon [5, 999] such that f is irreducible for some k epsilon [2, n 2], there are 317 values of n such that k epsilon (n/2, 2n/3]. For these irreducible trinomials, the space complexities of the CRT -based hybrid multipliers are reduced by 14.3% on average. As a comparison, the previous CRT -based multipliers considered the case k epsilon [2, n/2], and the improvement rate is 8.4% on average for only 290 values of n among these 539 values of n.
引用
收藏
页码:55 / 63
页数:9
相关论文
共 50 条