Variable length number chains generation without repetitions

Abstract

Pseudorandom and random numbers generators, plays an important role in solving many real or simulated problems, in different domains such as Scientific Computing, Physics, Chemistry, Computer Science, Artificial Intelligence, Chaos, Games theory, Statistics, Economics, etc. that directly or indirectly include a probabilistic element. These generators can be found in calculators, compilers, spreadsheets, electronics files or library tables, However, the progressive use of increasingly sophisticated models will demand a fast pseudorandom number generation process, which can generate strings of arbitrary sizes, and ensure it's reproducibility, uniformity and statistical independence, hence it constitutes an active research field area. This paper presents a novel method for obtaining these numbers relevant to various branches of computational optimization. © 2010 Springer-Verlag Berlin Heidelberg.