RF sensor for non-invasive cardiopulmonary monitoring

Cardiopulmonary information can be extracted from the temporal variations of the input reflection coefficient of a single wearable antenna placed in close proximity of the human thorax. In a previous paper, the authors have shown the potentials of such non-invasive measurement technique through experimental results; as a proof of concept, phase samples were collected by using a Vector Network Analyzer, and conventional non-linear filtering techniques were used to isolate the spectral components related to heartbeat and respiration rate. To get more realistic measurement data, a first prototype of a low-cost RF sensor has been implemented, and improved algorithms have been developed to estimate both heartbeat and breathing rate. Preliminary measurement results are shown to validate the approach, and the effects of the human body movements are discussed.