The non-ideal properties of a filter lead to channelization distortion, which affects signal detection. In order to achieve high precision signal parameters estimation and waveform recovery accurately, a method of channelized detection of WLFM signal based on basis pursuit is presented. The method uses l1 norm sparse regular least squares model and deduces the SOCP form of the model. It expands the output signals of each subband over complete Gabor atom dictionary and achieves the parameter estimation and decomposition and reconstruction of the original signal. Simulation results show that the proposed method realizes the accurate time-frequency analysis and parameter estimation of the signal, reconstruct original wide band signal through sparse atoms (less data) and reduces the impact of channelization distortion to some extent.