一种基于多目标差分演化的序列密码算法  

Sequence encryption based on multi-objective differential evolutionary algorithm

作　　者：吴君钦[1] 陈天栋 李康顺[1,2] 

机构地区：[1]江西理工大学信息工程学院,江西赣州341000 [2]华南农业大学信息学院,广州510642

出　　处：《计算机应用研究》2014年第7期2139-2143,2147,共6页Application Research of Computers

基　　金：国家自然科学基金资助项目(70971043);江西省教育厅科学技术研究项目(GJJ12348;GJJ12368)

摘　　要：为了避免传统序列密码算法中出现的生成序列密码重码率高和容易陷入局部最优解等缺点,提出了一种基于多目标差分演化的序列密码算法(DEMOSEP)。该算法将用于评价序列随机性的三个指标——频数检验、序列检验和游程检验作为优化的三个目标函数,同时利用差分演化算法的高效且全局搜索能力的操作算子,并融合快速非支配排序和基于拥挤距离的选择机制,智能地演化出满足这三个指标的序列密码。实验结果表明,利用该算法产生的序列密码能够较好地通过各项随机性检验,相比传统演化算法生成的序列,具有更高的随机性和安全性,并具有一定的实用性。To avoid the disadvantages of traditional sequence ciphers algorithm with highly coincident code rate and easily falling into local optimal,this paper proposed a sequence encryption generating method based on multi-objective differential evolutionary algorithm（DEMOSEP）. The presented algorithm used three indicators as the optimized objective functions,namely frequency-test,sequence-test and runs-test to evaluate the randomness of sequence. In addition,it used the efficient operators of differential evolution,and combined the fast non-dominated sorting with the crowded distance selection mechanism to evolve sequence ciphers which could meet these three indicators. The experimental results show that the sequence encryption generating method has high randomness and security,and also has some practicality.

关 键 词：序列密码 随机性 差分演化算法 多目标优化 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]