Abstract: The relation between sub-apertures configurations of optical sparse-aperture imaging systems and MTF is discussed. A new method, based on Monte Carlo inversion, for the aperture configuration optimization of sparse-aperture systems is proposed. The number of the aperture models under test is greatly reduced through discrete model space and optimization speed is improved through searching the model space by alternate random search and systematic search. The effectiveness of the new method was evaluated and analysed by computer simulation experiment. The optimality results of the three, four, five, six and seven sparse-aperture structures are accordance with the results of the other references. The optimization efficiency of the method was tested simultaneously. The test result shows that the method has higher optimization efficiency.