Abstract: The coherent accumulation problem of distributed multiple-input multiple-output (MIMO) radar was studied. We proposed a phase compensation and coherent accumulation algorithm of multi-node echoes, which used information entropy as an optimization criterion. Orthogonal frequency division multiplexing linear frequency modulated signal (OFDM-LFM) was used as emission waveforms. The signal separation of coupling echoes and pulse compression were completed by matched filtering. In order to eliminate the impact of step frequency, we took frequency shift operation in the frequency domain. Information entropy function of the range alignment echoes was used to estimate the phase parameters, then we constructed the phase compensated function to compensate the phase differences of multi-node echoes. Finally, We obtained the coherent accumulation result of distributed MIMO radar. The proposed method can deal with the echoes of multiple nodes at the same time, simulation results are proided to verity the validity of the proposed method.