A Novel Feedback Block Cipher Based on the Chaotic Time-Delay Neuron System and Feistel Network

In this paper, we propose a novel feedback block chaotic cipher algorithm (FBCCA) based on the chaotic time-delay neuron system and feistel network. The inherent features of chaotic system make the cryptosystem more complex and more difficult to be analyzed or predicted. In FBCCA, the output ciphertext block corresponding to the current input plaintext block is fed back into the generation process of the number of round, which will be used for encrypting the next plaintext block. Furthermore, the round numbers (times iterated for each block) and the sequence number of S-box (used in each round) have a closely relationship with the secret key, and both are dynamically generated by chaotic system. The results of computer simulation experiments show that the proposed cipher has excellent cryptographic properties, i.e., the cipher is very sensitive with respect to the slight change of the plaintext and the secret key, and the randomicity of the ciphertext is very ideal. At last, the security of the algorithm is studied, and the results indicate that the algorithm possesses higher capability of resisting the brute-force attack. Some further studies are also suggested in this paper.