Fast selection of time-interleaved samples for wireless healthcare monitoring with pulse radar

In this paper, we exploit using ultra-wide band(UWB) pulse radar for non-contact healthcare monitoring such as respiration or heartbeats rates. The transmitter sends periodic Gaussian pulses towards the person under monitoring. The pulses arrive at the person and reflect back to the receiver. The receiver samples the reflected signal in RF domain directly by time-interleaved sampling. The 16 time-interleaved analog-to-digital converters (ADC) lead to equivalent sampling rate of 20-GSamples/s. The respiration and heartbeats information are captured in the reflected signal and processed by the system. We have developed a successive interference cancelation algorithm at the receiver to post-process the signals and extract the respiration rate and the heartbeats rate successively. Furthermore, instead of taking all the received samples, a low complexity scheme is developed to select from the 16 interleaved ADC subchannels. We find that only a subset of the 16 interleaved samples are sufficient to maximize the signal-to-noise ratio. Also,the proposed scheme does not require prior knowledge of the pulse waveform. The performance evaluation based on the mean square error (MSE) of estimated heartbeats rate is also presented.