An Efficient PRNG Based on the Hybrid between One- and Two-Dimensional Cellular Automata

This paper proposes an efficient pseudorandom number generator (PRNG) based on hybrid between one- and two-dimensional cellular automata (CA). The proposed CA PRNG consists of two phases according to dimension. The first phase is the evolution of CA cells in two-dimension (2- D) with a von Neumann neighborhood. To achieve a high quality of randomness, a distinguishable method from previous CA PRNGs is chosen. The second phase is the evolution of CA cells in one-dimension (l-D) with 2-state, 3- neighborhood. In the evolution of l-D CA, <90,150> rule combination is used. To evaluate the quality of randomness, the ENT and DIEHARD test suites are used. The results of these tests have shown that the high quality of randomness is better than previous proposed CA PRNGs.