A new class of efficient piecewise nonlinear chaotic maps for secure cryptosystems

Abstract

In this paper we construct a new class of nonlinear chaotic maps for secure cryptosystems. These maps can overcome the security holes brought by the "piecewise linearity" of the previous Piecewise Linear Chaotic Maps (PWLCM) due to a fact that the chaotic sequences generated by the derived iterative system based on the proposed maps are proved to have perfect dynamic properties, such as uniform invariant distribution, d-like autocorrelation function etc. Furthermore, the relative quantized two-value sequences also have perfect secure statistical characteristics. In terms of computing speed, the proposed maps have faster speed than the recently proposed nonlinear "piecewise-square-root" maps (PSRM), and they actually have equivalently the same computing speed with the linear PWLCM.