In this digital signal processing procedure, which analyses Electroencephalogram (EEG) data within a specified framework, the signal's frequency sub-bands were broken down using DWT, and a collection of features representing the distribution of wavelet coefficients was retrieved from the sub-bands. With the use of feature extraction techniques like mean, standard deviation, and variance, the size of the data is decreased. Following that, the classifier utilised these attributes as input to determine if the input data was normal or drowsy, and it classified the data accordingly. To demonstrate the superiority of the classification process, the performance of classification is evaluated. The system's main goal is to increase the precision of patient monitoring systems, remove variations in EEG signals, and enhance process accuracy.