Including the magnitude variability of a signal into the ordinal pattern analysis

TL;DR

This paper enhances ordinal pattern analysis by incorporating signal magnitude variability, improving the characterization of dynamical behaviors and sleep-wake states, and aiding feature engineering for AI classifiers.

Contribution

It introduces a novel method to include magnitude variability into ordinal pattern analysis, addressing the loss of magnitude information in traditional OP encoding.

Findings

Improved differentiation of dynamical behaviors in synthetic maps.

Enhanced classification of sleep-wake states using combined permutation entropy and magnitude variability.

Potential for better feature extraction in machine learning applications.

Abstract

One of the most popular and innovative methods to analyse signals is by using Ordinal Patterns (OPs). The OP encoding is based on transforming a (univariate) signal into a symbolic sequence of OPs, where each OP represents the number of permutations needed to order a small subset of the signal's magnitudes. This implies that OPs are conceptually clear, methodologically simple to implement, robust to noise, and can be applied to short signals. Moreover, they simplify the statistical analyses that can be carried out on a signal, such as entropy and complexity quantifications. However, because of the relative ordering, information about the magnitude of the signal at each timestamp is lost -- this being one of the major drawbacks in the method. Here, we propose a way to use the signal magnitudes discarded in the OP encoding as a complementary variable to its permutation entropy. To illustrate our approach, we analyse synthetic trajectories from logistic and H{\'e}non maps -- with and without added noise -- and intracranial electroencephalographic recordings from rats in different sleep-wake states. Our results show that, when complementing the permutation entropy with the variability in the signal magnitudes, the characterisation of the dynamical behaviours of the maps and the sleep-wake states is improved. This implies that our approach can be useful for feature engineering and improving AI classifiers, where typical machine learning algorithms need complementary signal features as inputs to improve classification accuracy.