High-speed low-power parallel Random Number Generator for wireless channel emulators

Uniform Random Number Generators (URNG) are an indispensable block in channel emulators systems. They are required for producing noises and distortions for testing wireless communication systems. Therefore, there is a necessity of producing several uncorrelated URNG. This paper presents a high-speed, low-power URNG generator for producing parallel uncorrelated data flows, based on the use of the hybrid cellular automata´s (CA) method. Time and area analysis carried out, showing that the proposed architecture can be used in parallel to test multiple-input multiple output (MIMO) systems. Statistical analyses are presented corroborating its quality. In addition, the proposed architecture has been implemented in a VLSI device, showing its potentially massive use in commercial channel emulators.