Sleep stage classification using time-frequency distribution of bioradiolocation

Jemali, Muhamad Adam (2021) Sleep stage classification using time-frequency distribution of bioradiolocation. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Sleep stage classification using time-frequency distribution of bioradiolocation.pdf - Submitted Version Download (1MB)
	Text (Full text) Sleep stage classification using time-frequency distribution of bioradiolocation.pdf - Submitted Version Restricted to Repository staff only Download (2MB)

Abstract

The quality of life is significantly affected by sleep problems. The categorization of sleep stages and method used in quantify the quality of sleep remains unresolved question in the area of sleep research. The time-domain and frequency-domain based methods are not appropriate for sleep stage categorization due to the presence of low frequency signals. This study therefore was carried out to classify and evaluate sleep stages using Bio-radiolocation (BRL) data based on time-frequency distribution (TFD) method and compare with different classifier. A total of 16 signals data from 8 samples were segmented into 4 sleep stages using time-frequency spectrum as RMS voltage. The sleep stage classification was then compared with the different classifiers of Simple Vector Machine (SVM), K-nearest neighbor (KNN) and Ensemble classifiers. Results showed that the KNN model demonstrate the highest accuracy of 96.9% in classifying the sleep stages using TFD method compared to other classifiers. The experimental results presented from this study showed the proposed approach is effective in increasing the accuracy of the sleep stage detection using BRL data.

Actions (login required)

View Item