By comparing the development of data assimilation in meteorology, Optimal Interpolation (OI) assimilation in one-dimension ionosphere is introduced. OI assimilation is a method that minimizes error variance of analysis by non-linear least-squares. Based on IGGCAS1D (Institute of Geology and Geophysics, Chinese Academy of Sciences 1 Dimension) model and Xiamen digital ionosonde observations, OI assimilation has been validated by simulation experiments. The detailed information of observations is that, 11:00UT 15 June, 11:00UT 22 July, 11:15UT 31 August, 2009, respectively. The results show that analysis conforms to observation even if the model background is far away from it and that OI assimilation is validated. After that, using one of the results from simulation experiments (11:15UT, 31 August 2009), a residual prediction is introduced. The result of residual prediction accords with observation very well. From the difference between prediction and observation, it can be concluded that the residual prediction is a worthy method to be verified in future.