Abstract: he MIMO radar system consists of a transmit array with widely spaced elements such that each element views a different aspect of the target.The array at the receiver is a conventional array used for direction finding.MIMO radars combine both the advantages of multiple signals technology and the array technology，so as to decrease the influence of target RCS(radar cross­section)scintillation owing to the space diversity when detecting and tracking targets.For target detection of MIMO radars，we put forward a distributed fusion detection method by non­coherent accumulation based on the Markov chain.Firstly，we analyze the calculation principle for the threshold of constant false alarm rate(CFAR)detection，then deduce secondary detection threshold after multiple station data fusion.A fusion detection method based on Markov chain is leveraged toreduce the calculation burden of Bayesian and Newman­Pearson detection，.The proposed method structuralizes the echo data and introduces Markov process to calculate the state transition probability matrix according to the state statistical relationship between the different time，then searches target position by calculating the echo data in sequence.The simulation results show that the proposed algorithm can distinctly improve the detection rate without increasing the false alarm probability.Thus the system detection performance can be effectively improved.