For the nonlinear, non-stationary and weak radar life signal with serious noise interference and multiple singularity points, denoising is the essential means before analyzing the useful signal. The radar life signal model is established based on the principle of Doppler effect and the statistical characteristics of noise. Wavelet transform and lifting wavelet transform are applied respectively on the radar life signal under strong noise background for denoising. The results show that for radar life signal under strong noise pollution, it is an efficient denoising method to use wavelet transform or lifting wavelet transform. The lifting wavelet transform denoising effect is better than traditional denoising methods of wavelet transform. The signal-to-noise ratio (SNR) and mean square error (MSE) are higher than those of traditional wavelet denoising. In the radar life signal denoising, the wavelet basis function is sym8 and the number of decomposition layers is 3.