Fast Chaotic Image Encryption with Simultaneous Permutation and Diffusion for IoT Applications

Abstract

The current prevalence of the internet for secret communication, alongside the development of the Internet of Things (IoT), increases the importance of cryptography for information security. Considering that the most widely shared archives are images and that traditional encryption algorithms, such as the Advanced Encryption Standard (AES), are not optimized for image encryption, researchers have investigated the use of chaos theory in image encryption schemes. However, their elevated encryption time and memory access, due to the number of scannings needed to cipher an image, have been an issue for chaotic ciphers. Considering this, a new cipher is proposed in this work in which the plain-image is divided into blocks and a simultaneous permutation-diffusion is adopted, increasing the cipher's time efficiency. The cipher is executed in a Raspberry Pi 3 device, and it presents a significant increase in throughput compared to other similar works in literature. Additionally, extensive security evaluations showed that the cipher is secure relative to statistical, differential, chosen-plaintext, brute force, noise, and occlusion attacks, which are hence suitable for the application in real-time IoT applications.