Radar-based measurements of periodic movements

In this work, the continuous wavelet transform (CWT) is used to analyze periodic motions, measured by contactless Doppler radar, which have small amplitudes, drifting frequencies and dropouts. Using a classical method like fast Fourier transform (FFT) for periodic signals results in loss of the time behavior of the signal, drifting frequencies and dropouts, causing distortion in the frequency peaks of the FFT signal. An alternative is the short-time Fourier transform (STFT), also known as ldquowindowed FFTrdquo (WFFT), where additional time information is obtained by sequential time shifted windowing of the input signal, but the fixed window size limits its time-frequency resolution. The wavelets of the CWT have frequency dependent sizes, permitting the CWT to have an optimal time-frequency resolution. This enables the identification and tracking of the frequency, as well as detecting dropouts of the signals.