A novel simultaneous permutation-diffusion image encryption scheme based on a discrete space map

The constant increase of devices connected to the internet and the development and spread of technologies such as 5G leads to substantial growth in data-sharing over non-secure channels in real-time applications, augmenting the need for fast cryptography algorithms. Algorithms such as the Advanced Encryption Standard (AES) are not optimized for image encryption, and hence, as an alternative, chaotic encryption strategies have been explored by researchers. However, due to the use of floating-point arithmetic and inefficient memory access, the performance of these chaotic ciphers is insufficient for the encryption of high-resolution images. Thus, this work proposes generating the chaotic sequences based on fixed-point arithmetic to increase computational efficiency, and using a simultaneous permutation-diffusion operation and block encryption to deal with the inefficient memory access. The proposed scheme leads to a throughput 353 times higher than the AES, 75 times higher than the previously higher value in the literature and 11 times higher than the Ascon algorithm. The cipher security evaluation is performed based on statistical attacks, with metrics such as Pearson, histogram variance, Shannon's entropy, correlation coefficients, and NIST tests; differential attacks, with the NPCR, UACI, and MSE metrics; chosen plain-text, with the SSIM metric; brute-force, with the key-space; noise and occlusion attacks, with the PSNR metric.