On-Body Continuous Wave Radar (CWR) systems can be used to monitor cardiac activities. This paper presents a signal processing method based on the wavelet transform to remove the noise components of the time and frequency domains of CWR signals obtained from an on-body radar system in order to derive heart rate more accurately. The proposed method is evaluated with CWR signals recorded from the human body in two different postural and exercise activities. The accuracy and precision of the suggested method are evaluated by comparing the heart rate derived from a reference Electrocardiogram (ECG) signal. © 2018 IEEE.