A multiscale Kalman filter based on wavelet is proposed for signal restoration embedded in 1/f fractal noise. First,the signal with 1/f fractal noise is transformed into multiscale subsystems in time-scale domain. Some nonstationary properties of fractal signals are attenuated in each subband by wavelet multiresolution decomposition,so that the Kalman filter bank can be applied to estimate the multiscale input signals. Then the estimated multiscale input signals are synthesized to obtain the estimated signal. Some simulation examples are given for testing the performance of the proposed algorithm and output signal to noise ratio is improved evidently.