Abstract
Heartbeat interval time series is an example of natural signals with 1/f characteristics. The exponent α of the 1/fα spectrum has some clinical significance. But sometimes the 1/f components is superimposed by some sinusoid components in the signal. To estimate the slope accurately, the 1/f component must be extracted from the signal. The singular spectrum analysis (SSA) method is recruited here to perform the task. Experimental results on data from real patients are satisfactory.
Original language | English |
---|---|
Pages (from-to) | 302-304 |
Number of pages | 3 |
Journal | IEICE Transactions on Information and Systems |
Volume | E83-D |
Issue number | 2 |
State | Published - 2000 |
Externally published | Yes |
Keywords
- 1/f characteristics
- Heart rate variability
- Singular spectrum analysis