Kiyoshi Kudo

Mutual information of ising systems

We obtain the mutual information of Ising systems, which shows singular behavior near the critical point. We connect the mutual information with the magnetization and the correlation function. The mutual information is a suitable measure for the critical behavior of Ising systems.

Edge of chaos in rule-changing cellular automata

Abstract We present a model of rule-changing cellular automata (CA) to study dynamical structure of CA in detail. In our model calculation the edge of chaos is observed on the basis of the source rule 22 which yields the most interesting behavior. We find that the edge of chaos is generated through competition between two rule-changing behaviors. These rule-changing behaviors are analyzed quantitatively by using the mutual information as a measure ofcomplexity. It is noted that the mutual information has the maximum value near the edge of chaos because of competition.

A complexity measure of the Internet

The diversity of entropy, as a measure of complexity, for data traffic on Internet has been studied. We also calculated the measure of artificial sequences, such as periodic, random, mixed, cellular automata generated and logistic map generated sequences, to compare with that of data traffic on the Internet. As a result, we find, the measure of data traffic is larger than those produced by other artificial sequences.

Complexity and Diversity for the Rule Changeable Elementary Cellular Automata

The rule changeable elementary cellular automata are proposed analogous to the ecological system. In this model we find the behavioral diversity for a suitable threshold value x on chaotic source rule. The behavioral diversity is characterized by block entropy and joint entropy. Our behavioral diversity plays just as “edge of chaos” for spatial and temporal evolution.

Dynamics of two coupled inverted pendulums

Dynamics of two coupled van der Pol type inverted pendulums are studied. A phase transition is observed between synchronized phase and asynchronized phase. Behavior of angular correlation near critic2Ll point is observed. Exact critical point and critical exponent are obtained by obeserving convergent time.

Analysis of concentration of atmospheric radon daughters as a time series data

A relation of radon (Rn) daughters concentration in the air to meteorological parameters is investigated. The atmospheric radon daughter concentration and the related meteorological parameters are treated as a simple time series data. The power sprctrum of radon daughter concentration and temperature after fourier transfer are fit to 1/f closely. This fact confirms that the common dynamics exist between the variations of the parameters.