将混沌机制的非线性和随机特性引入密码系统 ,是加强信息保密的有效途径。离散 Hopfield神经网络存在混沌吸引子 ,基于此可以构造新的加密算法。该文在 GuoDonghui等人提出的对称的概率加密算法的基础上利用密码分组链接模式 (CBC) ,构造了一个较为完整的神经网络分组密码系统 ,它与数字加密标准 (DES)相比 ,更为简单 ,并具有抵御一些类型攻击的优点 Introducing the chaos mechanism’s non-linearity and stochastic character into the cryptosystem is an effective way to enhance the information confidentiality. Discrete Hopfield neural networks have chaotic attractors, based on which new cryptographic algorithms can be constructed. Based on the symmetric probabilistic encryption algorithm proposed by Guo Donghui et al., This paper constructs a more complete neural network block cipher system based on the CBC, which is more in line with the Digital Encryption Standard (DES) Simple and has the advantage of being resistant to some types of attacks