Singular Spectrum Analysis Window Length Selection in Processing Capacitive Captured Biopotential Signals

Biopotential signals captured by non-contact capacitively coupled biopotential sensing electrodes are easily polluted by different kinds of noises, especially power line noises. Singular spectrum analysis (SSA) algorithm is utilized for processing capacitively captured biopotential signals. By analyzing the spectrum characteristics of the circulant matrix and Toeplitz matrix, it has been found that the window length of SSA is strongly related to the frequency width corresponding to each principal component. A criterion for window length selection is, therefore, proposed and numerically verified by simulated electrocardiography signals. A capacitive-coupled biopotential signal measurement system is constructed to obtain original biopotential signals from human body. Experimental results show that, with the aid of the criterion, satisfied biopotential signals can be effectively obtained by SSA.