The invention discloses a pedestrian indoor positioning method based on step matching, which relates to the indoor positioning technology field. The invention mainly aims at the solution to the problem that in the study of the indoor positioning pedestrian dead reckoning (PDR) algorithm, the traditional step model does not fully reflect the individual differences and therefore, puts forward a method based on the satellite ranging and establishing an individual step characteristic database to replace the step model method. In the invention, a smart phone is used as the carrier so that the satellite ranging outdoors is applied to the measurement of step length, and a corresponding relation between different velocity characteristics and step length is established. In the room, the improved K-nearest neighbor (KNN) algorithm matches the step length in real time and calculates the current position of the pedestrian by combining the direction information. Aiming at individual differences, the invention no longer adopts a single step model, and the indoor positioning accuracy is higher than that of the traditional step model.