Processing Binary and Fuzzy Logic by Chaotic Time Series Generated by a Hydrodynamic Photochemical Oscillator

This work demonstrates the computational power of a hydrodynamic
photochemical oscillator based on a photochromic
naphthopyran generating aperiodic time series. The chaotic
character of the time series is tested by calculating its largest
Lyapunov exponent and the correlation dimension of its attractor
after building its phase space through the Takens’ theorem.
Then, the chaotic dynamic is shown to be suitable to implement
all the fundamental Boolean two-inputs-one-output logic
gates. Finally, the strategy to implement fuzzy logic systems
(FLSs) based on the time series is described. Such FLSs promise
to be useful in the field of computational linguistics, which is
concerned with the development of artificial intelligent systems
able to transform collections of numerical data into natural
language texts.