acquiring channel data of a terminal;

constructing a first feature vector based on the channel data, wherein the first feature vector represents a numerical value set of cross-correlation values, which change along with time intervals, between channel data at different moments and the channel data themselves and between the channel data at different moments and subsequent channel data at different time intervals; and

inputting the first feature vector into a predetermined prediction model to predict a speed of the terminal, or to predict a speed type of a cluster to which the terminal belongs;

wherein before inputting the first feature vector into the predetermined prediction model, the method further comprises:

establishing a model library, wherein the model library stores prediction models under different frequency points; and

selecting a prediction model corresponding to a frequency point of the terminal from the model library as the predetermined prediction model.