基于细胞神经网络的伪随机数生成方法

doi:10.11959/j.issn.1000-436x.2016252

摘要/Abstract

摘要：

为了克服有限精度效应对混沌系统的退化影响，改善所生成随机序列的统计性能，设计了一种新的基于六维CNN（细胞神经网络）的64 bit伪随机数生成方法。在该方法中，通过控制六维CNN在每次迭代过程中的输入输出，改善了混沌退化对随机数的性能影响，同时，通过与Logistic映射所生成的随机序列和可变参数进行异或处理，有效避免了生成序列的重复出现，扩大了密钥空间和输出序列的周期。以新方法设计的PRNG（伪随机数生成器）易于在软件中实现，每次可生成64 bit的伪随机数，生成速率快。测试结果表明，该方法生成的伪随机序列可以完全通过随机数检测标准NIST SP800-22，因而具有很好的随机性，可用于保密通信等信息安全领域。

关键词: 混沌系统, 6-CNN, Logistic映射, PRNG

Abstract:

To overcome the degradation characteristics of chaos system due to finite precision effect and improve the sta-tistical performance of the random number,a new method based on 6th-order cellular neural network (CNN) was given to construct a 64-bit pseudo random number generation (PRNG).In the method,the input and output data in every iteration of 6th-order CNN were controlled to improved the performance of the random number affected by chaos degradation.Then the data were XORed with a variable parameter and the random sequences generated by a Logistic map,by which the repeat of generated sequences was avoided,and the period of output sequences and the key space were expended.Be-sides,the new method was easy to be realized in the software and could generate 64 bit random numbers every time,thus has a high generating efficiency.Test results show that the generated random numbers can pass the statistical test suite NIST SP800-22 completely and thus has good randomness.The method can be applied in secure communication and other fields of information security.

Key words: chaotic systems, 6-CNN, Logistic map, PRNG

董丽华,药国莉. 基于细胞神经网络的伪随机数生成方法[J]. 通信学报, 2016, 37(Z1): 85-91.

Li-hua DONG,Guo-li YAO. Method for generating pseudo random numbers based on cellular neural network[J]. Journal on Communications, 2016, 37(Z1): 85-91.

图/表 7

图1

图2

表1

表2

图3

图4

表3

参考文献 21

[1]	KILLMANN W , SCHINDLER W . A design for a physical RNG with robust entropy estimators[C]//International Workshop on Crypto-graphic Hardware and Embedded Systems. Springer Berlin Heidelberg8, 2008:146-163.
[2]	MA Y , LIN J CHEN T , et al. Entropy evaluation for oscillator-based true random number generators[C]//International Workshop on Cryp-tographic Hardware and Embedded Systems. Springer Berlin Heidel-berg, 2014:544-561.
[3]	KELSEY J , MCKAY K A TURAN M S . Predictive models for min-entropy estimation[C]//International Workshop on Cryptographic Hardware and Embe dded Systems. Springer Berlin Heidelberg, 2015:373-392.
[4]	杨自强，魏公毅 . 综述：产生伪随机数的若干新方法[J]. 数值计算与计算机应用， 2001,22(3):201-216. YANG Z Q , WEI G Y . Review:several new methods of generating pseudo random numbers[J]. Numerical Calculation and Computer Ap-plications, 2001,22(3):201-216.
[5]	GU X C . Multi-output LFSR based uniform pseudo random number generator[J]. Geomatics ＆ Information Science of Wuhan Univer-sity, 2010,35(5):566-569.
[6]	FOKHRULISLAM M , A M A M , YEOP M B . FPGA implementation of an LFSR based pseudo random pattern generator for MEMS test-ing[J]. International Journal of Computer Applications, 2013,75(11):30-34.
[7]	SINGLA J W , SACHDEVA P , AHMAD M . A chaotic neural network based cryptographic pseudo-random sequence design[C]//2014 Fourth International Conference on Advanced Computing ＆ Communication Technologies. 2014:301-306.
[8]	WANG X Y , QIN X . A new pseudo-random number generator based on CML and chaotic iteration[J]. Nonlinear Dynamics, 2012,70(2):1589-1592.
[9]	FRANCOIS M , GROSGES T , BARCHIESI D , et al. A new pseudo-random number generator based on two chaotic maps[J]. In-formatica, 2013,24(2):181-197.
[10]	LUI O Y , YUEN C H , WONG K . A pseudo-random number generator employing multiple rényi maps[J]. International Journal of Modern Phys ics C, 2013,24(11).
[11]	WANG X , LIU L . Cryptanalysis of a parallel subimage encryption method with high-dimensionalchaos[J]. Nonlinear Dynamics, 2013,73(1-2):795-800
[12]	齐迎宾，孙克辉，王会海，等. 超混沌伪随机序列生成器设计与性能分析[J]. 计算机工程与应用， 2015. QI Y B , SUN K H , WANG H H , et al. The design and performance analysis of hyper-chaotic pseudo-random sequence generator[J]. Computer Engineer and Applications, 2015.
[13]	许栋，崔小欣，王田，等. 基于Logistic映射的混沌随机数发生器研究[J]. 微电子学与计算机， 2016(2):1-6. XU D , CUI X X , WANG T , et al. The study of chaotic random number generator based on Logistic map[J]. Microelectronics ＆ Computerh, 2016(2):1-6.
[14]	姚勇 . 熵、分维、李雅普诺夫指数与混沌[J]. 自然杂志， 1987(5):41-47. YAO Y . Entropy,fractal dimension,Lyapunov exponent and chaosl[J]. Nature, 1987(5):41-47.
[15]	CHUA L O , YANG L . Cellular neural networks:theory[J]. IEEE Transactions on Circuits ＆ Systems, 1988,35(10):1257-1272.
[16]	WANG X , BING X , ZHANG H . A multi-ary number communication system based on hyperchaotic system of 6th-order cellular neural net-work[J]. Communications in Nonlinear Science ＆ Numerical Simula-tion, 2001,15(1):124-133.
[17]	LI Y , XIAO D , LI H , et al. Parallel chaotic Hash function construction based on cellular neural network[J]. Neural Computing ＆ Applications, 2011,21(7):1563-1573.
[18]	张雪锋，范九伦 . 基于线性反馈移位寄存器和混沌系统的伪随机序列生成方法[J]. 物理学报， 2010,59(4):2289-2297. ZHANG X F , FAN J L . The method of generating pseudo random se-quence based on LFSR and chaos[J]. Acta Physica Sinica, 2010,59(3):2289-2297.
[19]	RUKHIN A , SOTO J , NECHVATAL J , et al. A statistical test suite for random and pseudorandom number generators for cryptographic ap-plications[J]. Andrew Rukhin Juan Soto James Nechvatal Miles Smid Elaine, 2010,59(4):2289-2297.
[20]	王亚红 . 基于 Henon映射参数扰动的伪随机数发生器设计及应用[J]. 中文信息， 2010,59(4):2289-2297. WANG Y H . The design and applications of PRNG based on Henon map with parameter perturbation[J]. Journal of Chinese Information Processing, 2010,59(4):2289-2297.
[21]	HOSSAIN M B , RAHMAN M T , RAHMAN B M S , et al. A new approach of image encryption using 3D chaotic map to enhance secu-rity of multimedia component[C]//International conference on Infor-matics, Electronics ＆ Vision. 2014:1-6.

密钥变化	位变化率
0x532a42d2→0x432a42d2	49.998 6%
0x78a12bc4→0x78a12bc5	50.002 6%

统计测试	P-value	Propotion	结果
Frequence	0.693 581	0.990 7	Pass
BlockFrequence	0.832 273	0.984 2	Pass
CumulativeSums^*	0.251 036	0.987 0	Pass
Runs	0.509 353	0.987 0	Pass
LongestRun	0.509 906	0.981 4	Pass
Rank	0.582 379	0.986 0	Pass
FFT	0.251 349	0.986 0	Pass
OverlappingTemplate	0.149 171	0.985 1	Pass
Universal	0.703 826	0.987 9	Pass
LinearComplexity	0.122 987	0.988 8	Pass
ApproximateEntropy	0.150 973	0.987 0	Pass
Serial^*	0.376 366	0.984 2	Pass
NonOverlappingTemplate^*	0.214 120	0.983 2	Pass
RandomExcursions^*	0.301 571	0.985 2	Pass
RandomExcursionsVariant*	0.811 644	0.986 7	Pass

方法	信息熵
本文方法文献[20]方法	7.997 4
	7.989 1
文献[21]方法	7.989 0