In order to reduce the measurement error of the displacement sensor ba sed on self-mixing interference,a signal processing method based on wavelet analysis is proposed.Before the all-phase spectrum ana lysis,wavelet analysis is applied to remove the singular points of normalized signals.This method can reduce the impact of higher harmon ic produced by the slope mutation of modulation current and diminish the disturbing of high-frequency noise very effectively.F irst,the principle of displacement measurement based on self-mixing interference is described briefly.Second,the signal processing algorithm is i nvestigated with arithmetic simulation.Third,the influence of feedback level a nd modulation parameter on the error of displacement is analyzed by numerical si mulation.At last,experiment system is established and the calibration result of piezoelectric ceramics is presented.Experimental results indicate that the displacement error can be decreased to 2.45nm,so the wavelet analysis can r educe the displacement error of self-mixing interferometer.