A Light-weight ESalsa20 Ciphering based on 1D Logistic and Chebyshev Chaotic Maps

Abstract

Information security has become a significant component in Information and Communication Technology (ICT). Different ciphering methods have been introduced as a solution to supply confidentiality and act a greater role in systems of information security. The majority of these methods devour a lot of computing resources, for example, time of CPU, memory, and force. Salsa20 is a modern and efficient stream encryption algorithm and its versions have the fastest stream codes available today. Instead of the broken Salsa20/7 and the not safe method Salsa20/12, the characterizes of Salsa20 are depicted by the execution of all rounds of ciphering to obtain a high diffusion to avoid the known attacks. In this paper, a method for enhancing the Salsa20 algorithm (ESalsa20) is proposed that uses a chaotic layer to achieve the diffusion faster than the traditional Salsa20. The results of histogram analysis, comparison between proposal Esalsa20 and original Salsa20 based on the results of Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Average Difference (AD), Universal Image Quality (UQI), Normalized Cross-Correlation (NCC), Normalized Absolute Error (NAE), and execution time showed that the proposal improved ESalsa20 has less complexity and more secure than the original Salsa20. According to the obtained results, it is observed the proposed model is faster than the original algorithm, and it is achieved the best value for all experiments, the best values of NAD =0.3689770561, NCC=0.3890752653, and UQI=0.3777468183.

Keywords: Cryptography, stream cipher, Salsa, Chaos Theory,1D logistic function, Chebyshev function