华侨大学学报(自然科学版)
華僑大學學報(自然科學版)
화교대학학보(자연과학판)
JOURNAL OF HUAQIAO UNIVERSITY(NATURAL SCIENCE)
2014年
5期
520-523
,共4页
蔺小梅%李国刚%张泽普
藺小梅%李國剛%張澤普
린소매%리국강%장택보
序列密码%线性反馈移位寄存器%离散型 Hopfield神经网络%混沌吸引子
序列密碼%線性反饋移位寄存器%離散型 Hopfield神經網絡%混沌吸引子
서렬밀마%선성반궤이위기존기%리산형 Hopfield신경망락%혼돈흡인자
stream cipher%linear feedback shift register simulated%Hopfield neural networks%chaotic attractors
结合反馈型离散 Hopfield神经网络(HNN)和改进型的线性反馈移位寄存器(M?LFSR)的优点,提出基于 OHNN和 M?LFSR的字序列密码。该方案利用离散 Hopfield神经网络的混沌吸引子对改进型线性反馈移位寄存器的非线性选择输出,实现加密。安全性分析与仿真验证表明:该算法构造的伪随机序列具有良好的随机性、复杂度等特点,满足密码学的要求。
結閤反饋型離散 Hopfield神經網絡(HNN)和改進型的線性反饋移位寄存器(M?LFSR)的優點,提齣基于 OHNN和 M?LFSR的字序列密碼。該方案利用離散 Hopfield神經網絡的混沌吸引子對改進型線性反饋移位寄存器的非線性選擇輸齣,實現加密。安全性分析與倣真驗證錶明:該算法構造的偽隨機序列具有良好的隨機性、複雜度等特點,滿足密碼學的要求。
결합반궤형리산 Hopfield신경망락(HNN)화개진형적선성반궤이위기존기(M?LFSR)적우점,제출기우 OHNN화 M?LFSR적자서렬밀마。해방안이용리산 Hopfield신경망락적혼돈흡인자대개진형선성반궤이위기존기적비선성선택수출,실현가밀。안전성분석여방진험증표명:해산법구조적위수궤서렬구유량호적수궤성、복잡도등특점,만족밀마학적요구。
A new encryption technique in word?oriented stream cipher has been proposed.It consists modified linear feedback shift register simu?lated (M?LFSR),and a feedback discrete?time Hopfield neural networks (HNN).Nonlinear-ity of the M?LFSR has been chosen by the chaotic attractors of discrete?time HNN as the output to achieve encryption. Safety analysis and simulation have shown that the pseudo?random sequence constructed by the proposed algorithm has been characterized by good randomness and complexity,which meets the requirements of cryptography.