Deteksi Intensi Pergerakan Jari Menggunakan Metode Power Spectral Density dengan Stimulus Visual

Abstract

This research explores a detection of finger’s movement using Burg’s Power Spectral Density (PSD) as features vector. EEG signal is recorded using sampling frequency of 1000 Hz. Analysis of the signal is conducted by dividing signal into three segments; 1000 ms, 500 ms and 250 ms. Common Average Reference (CAR) and Support Vector Machine (SVM) are used in features extraction and pattern recognition. The result shows that the system can classify the finger’s movement with accuracy of about ±65,37% in 1000 ms of signal length.