Design and Implementation of a Configurable and Aperiodic Pseudo Random Number Generator in FPGA

Abstract

Pseudo random number generators (PRNGs) play an important role in many fields such as encryption and hardware testing. One common structure of PRNGs is linear feedback shifting register (LFSR), which generate a n-bit sequence from a specific seed. However, the sequence of random numbers can be predicted given a fixed seed and some sequences of generated numbers. Considering the drawbacks of available LSFR-based PRNGs, in this paper we devised a configurable and aperiodic LSFR-based PRNG, which utilize the unpredictability of metastable state widely existed in digital circuits. In this case, the sequence of random numbers is almost unpredictable and can meet the requirements of most applications especially critical ones. Then we implemented it on FPGA and the simulation results indicate that the proposed PRNG has better performance in generating random numbers with high randomness.