在预测步骤中，初始信息由观测数据的时间序列模型表示，下一次基于前一次数

In the prediction step, the initial information is represented by the time series model of the observation data, and the next time is based on the state of the previous data assimilation step. Update the step size to obtain the assimilation value using the observation data and the Kalman gain matrix

In the prediction step, the initial information is represented by the time series model of the observed data, and the next time is based on the state of the previous data assification step. The updated step uses observation data and the Kalman gain matrix to obtain assulation values

In the prediction step, the initial information is represented by the time series model of the observed data, and the next time is based on the state of the previous data assimilation step. The update step size uses the observation data and Kalman gain matrix to obtain the assimilation value<br>