Evolving cellular automata rules for multiple-step-ahead prediction of complex binary sequences

  • Authors:
  • A. V. Adamopoulos;N. G. Pavlidis;M. N. Vrahatis

  • Affiliations:
  • Medical Physics Laboratory, Department of Medicine, Democritus University of Thrace, GR-681 00 Alexandroupolis, Greece;Computational Intelligence Laboratory (CI Lab), Department of Mathematics, University of Patras Artificial Intelligence Research Center (UPAIRC), University of Patras, GR-26110 Patras, Greece;Computational Intelligence Laboratory (CI Lab), Department of Mathematics, University of Patras Artificial Intelligence Research Center (UPAIRC), University of Patras, GR-26110 Patras, Greece

  • Venue:
  • Mathematical and Computer Modelling: An International Journal
  • Year:
  • 2010

Quantified Score

Hi-index 0.98

Visualization

Abstract

Complex binary sequences are generated through the application of simple threshold, linear transformations to the logistic iterative map. Depending primarily on the value of its non-linearity parameter, the logistic map exhibits a great variety of behavior, including stable states, cycling and periodical activity and the period doubling phenomenon that leads to high-order chaos. From the real data sequences, binary sequences are derived. Consecutive L bit sequences are given as input to a cellular automaton with the task to regenerate the subsequent L bits of the binary sequence in precisely L evolution steps. To perform this task a genetic algorithm is employed to evolve cellular automaton rules. Various complex binary sequences are examined, for a variety of initial values and a wide range of values of the non-linearity parameter. The proposed hybrid multiple-step-ahead prediction algorithm, based on a combination of genetic algorithms and cellular automata proved efficient and effective.