The properties of geophones are important factors for high-resolution seismic exploration and have a great influence on data quality.High-resolution seismic exploration does not mean high-frequency exploration.If the natural frequency of the geophones is too high,not only the low-frequency energy will be suppressed and the frequency band will be reduced; but also the effective frequency band will fall into the nonlinear area,producing nonlinear distortion and lowering resolution.In this paper we combine the advantage of three kinds of different geophone frequencies high,low,and conventional to broaden the effective band width of the acquired data.For this purpose; we have designed a partial frequency band matched filtering method which can combine the advantages of low and high frequency to conventional frequency ranges.The technique of matched filtering involves exploiting the similarities of the unknown,weak signals and known stronger signals in the same area,supporting the low frequency signal.This process is modeled by introducing a richer wavelet with different dominate frequency and amplitude,by applying this filter to the conventional data; the effective band width has broadened and both resolution of the section and S/N ratio has been improved.