基于Hadoop大数据平台和无简并高维离散超混沌系统的加密算法

针对目前大数据环境中存在的数据安全问题,提出一种基于Hadoop大数据平台和无简并高维离散超混沌系统的加密算法.算法采用流密码对称加密方式,在Hadoop平台上读取存储于HDFS(Hadoop distributed file system)的大数据,进行分片处理和MapReduce编程后,用Map函数实现数据并行加密和解密,通过Reduce函数实现数据的合并操作并存储于HDFS.该算法具有较好的执行效率.与正李氏指数发生简并的低维混沌系统相比,无简并高维离散超混沌加密算法能提高系统安全性能,李氏指数均为正并且足够大,具有更好的统计特性,可通过严格的TESTU01测试,并行加密的密文之间互相关性很小.密钥参数众多使得估计或辨识难度增大.在密文闭环反馈条件下,具有抵御已知明文攻击和选择明文攻击的能力.

Encryption algorithm based on Hadoop and non-degenerate high-dimensional discrete hyperchaotic system

Aiming at the data security problem in big data environment, in this paper we propose a new chaotic encryption algorithm based on both big data platform named Hadoop and non-degenerate high-dimensional discrete hyperchaotic system. The algorithm utilizes the chaotic stream cryptography and reads the data from HDFS of Hadoop platform. After fragmentation processing and MapReduce programming, the data are encrypted and decrypted by Map function in parallel. The Reduce function implements the merging operation of the data and stores them on the HDFS. The algorithm has a better execution efficiency. Compared with the low-dimensional chaotic system based encryption algorithm, the non-degenerate high-dimensional discrete chaotic system based encryption algorithm can improve the system security performance. It can pass the strict TESTU01 test with better statistical properties and make sure that the correlation with the parallel ciphertext is very small. Numerous key parameters increase the difficulty in making estimation or identification. Under the closed-loop feedback in ciphertext, it has the ability to resist the known and chosen plaintext attacks.