The new infrared material (1-x)Pb(Mg1/3Nb2/3)O3-xPbTiO3(PMN-PT) has more sensitivities and noises because of its perfect piezoelectric and pyroelectric performances. The characteristic noises in the new infrared sensor are analyzed. We use the algorithm of auto-correlation and wavelet transform to analyze the model of readout signal with white noise and 1/f noise at 0 dB simulated by computer. The noise of actual signal is reduced by wavelet transform through field-programmable gate array (FPGA). The result shows that the auto-correlation can only reduce white noise, and the signal-to-noise ratio (SNR) of the signal after de-noise is improved by 5.5 dB. While the wavelet transform can reduce temperature noise, white noise and 1/f noise. The SNR of the signal after de-noise is improved by 15 dB.