EEG Signal Processing using Matlab | EEG Data Processing and Classification смотреть онлайн

Title: - Digital Signal Processing with EEG Signals
-----------------------------------------------

Implementation Plan:
---------------------

Step 1: Initially we load the EEG signals from EEG Motor Movement/Imagery Dataset.

Step 2: Next, we perform analyze the data, In this process we have Use the following methods, They are
2.1 Linear Method: The autoregressive moving average (ARMA) method and multivariate AR (MVAR) approach are used as linear methods to analyze EEG signals in time domain.
2.2 Non-linear Methods: The functions represent non-linearity in mean and variance of signal, respectively. In this method we used Short data sequences that are efficiently analyzed.
2.3 Short Time Fourier Transform: This method provides a bridge between the Fourier analysis and wavelet transform. The FT is not an appropriate method to be used when it comes to non-stationary signals. The segmentation of the signal is done using the window function that has equal width as the segment of the signal. The wavelet transform is preferred to be implemented for analyzing EEG signals because of its dual property i.e. it can be used for discrete (Discrete WT) and analog (Continuous WT).

Step 3: Next, we perform to reduce noises of EEG signals and find the correlation between signals (Channels) by using convolution theory.

Step 4: After completing the noise reduction process, we perform Separate the noise, In this step the reduced noise signals and normal noise signals are separated.

Step 5: After completing the Separate the noises step, we perform the Feature Extraction and classification process, The features such as standard deviation, mean value, variance, signal power, energy, min value, max value,variance. Then we classify the Features that are translated into commands that operate a device.

Step 6: Finally, we evaluate the following performance metrics

6.1 Number of epochs vs. training loss
6.2 Number of iterations vs. classification accuracy
6.3 F1 score
6.4 Precision
6.5 Recall
=======================================================================================================================================================

Software Requirement:
-----------------------

1. Tool: Matlab R2020a
2. Operating System: Windows 10 - (64-bit)
=========================================================================================================================================================