In the current paper a relatively simple phase-locked-loop (PLL)-algorithm based solution has been proposed in the few variants and evaluated on some real-time real-life signals. The proposed method can (in spite of the simplicity) be efficiently used in parallel with other methods, and taken into account if the PLL is clearly locked to the cardiac signal. Otherwise extraction results of other signal processing methods can be used for the system output. Also, as nowadays such PLL-based solution could be easily implemented in the software or firmware, also in embedded electronics solutions, advanced approaches can be simply used-like dynamic and adaptive on-line modification of the parameters of the PLL system, to improve the overall performance of the system. Also, as the locked PLL can easily provide synchronous reference signals, for the cardiac signal, and the cardiac signal is contains typically the components of fundamental frequency and harmonics of it, the cardiac signal can be easily decomposed (by simple synchronous detection) into the relatively sparse and compressed representation. Furthermore, the cardiac signal can reconstructed by simple means from such sparse representation. Breathing signal is relatively slow (low-frequency) signal and can be extracted (and analyzed and/or compressed, if needed) from the remaining signal, after extraction of the cardiac component.
