An indoor positioning method includes: obtaining, by a terminal, all predicted locations of a current location, predicting, according to historical movement information of the terminal, a first probability that the terminal passes through each predicted location at a next moment, and obtaining, based on strength of a received signal, a second probability that the terminal passes through each predicted location at the next moment; and generating a corresponding third probability according to the first probability and the second probability that are corresponding to each predicted location, and determining a predicted location point with a highest third probability as a location of the terminal at a second moment.