SCALING AND COMPLEXITY-ENTROPY ANALYSIS IN DISCRIMINATING TRAFFIC DYNAMICS

In this paper, the complexity-entropy causality plane approach is applied to analyze traffic data. The R/S analysis and detrended fluctuation analysis (DFA) methods are also used to compare with this approach. Moreover, based on the concept of entropy, we propose to use permutation to calculate the probability distribution of the time series when applying the representation plane. The empirical results indicate that traffic dynamics exhibit different levels of traffic congestion and demonstrate that this statistical method can give a more refined classification of traffic states than the R/S analysis and DFA.