A Comprehensive Survey on Detection of Non-linear Analysis Techniques for EEG Signal

This survey provides a comprehensive overview of the different methods for detecting the nonlinearity of the EEG signal. Electroencephalography (EEG) is a widely used signal for studying different brain functions. The nonlinearity of the EEG signal is often overlooked in many studies, but these nonlinear dynamics are important to understand and for characterizing the EEG. Different methods have been applied to determine the nonlinearity of the complex dynamics of EEG signals. In this paper, we review the current state-of-art chaos theories that can be applied to detect the nonlinearity of an EEG signal. The paper summarizes, in total, 28 research articles on six different methods: Correlation Dimension, Lyapunov Exponents, Recurrence Analysis, Surrogate Data Analysis, Nonlinear Regression, and Entropy-Based Measures. The review highlights the potential of nonlinear techniques for improving traditional EEG data analysis, the challenges of current nonlinearity detection techniques and the comparison between them, and the potential opportunities of these techniques in EEG signal analysis. With the findings from this survey, we propose a chaos-based nonlinear method that can overcome the challenges with current techniques and quantify the nonlinearity of all types of EEG signals. Overall, this survey serves as a resource on the latest nonlinearity detection techniques and their potential applications in EEG signal analysis.