A new approach of adaptive Kalman filtering deconvolution (AKFD) is developed based on dyadic wavelet transforms.The technique discards the assumption of signals stationarity in predictive deconvolution

and overcomes the problem of improving resolution at the price of substantially decreasing signal-to-noise rate (SNR).The technique can well compress the reflection waveforms

but the noises are not lifted in substance.So it has a better ability of noise tolerance.Suppressing false reflections in dyadic wavelet transform domain is better than by applying AKFD in the time domain.In addition

since the technique also has the characteristic of adaptive Kalman filtering in every band for a signal respectively

it enhances the adaptation of Kalman filtering

and the resolution being obvious higher than that in the time domain.At the same time

the technique also overcomes the drawback of increasing the low-frequency component of AKFD in the time domain.A great deal of numerical models and real seismic data indicate that the technique has obvious effects.