The invention relates to an indoor region class positioning method based on environment background sound sensing and analysis. Firstly, the environment background sound of all indoor regions is sensedthrough a smart phone microphone and is preprocessed; then, low-order acoustic fingerprints of sound signals are extracted; secondly, through a K-means clustering algorithm, an indoor space codebookset is constructed by combining the geometric contour information prior of the room, and high-order acoustic fingerprints of the sound signals are obtained; and finally, the high-order acoustic fingerprints are used as an input of a radial basis function core reverse propagation neural network, and through offline training and online test and verification, user position information is output. According to the method, due to the fact that the indoor space codebook set is introduced, the diversity of the indoor space low-order acoustic fingerprints is fully utilized, the accuracy of the positioning system can be improved, meanwhile, due to the fact that high-order acoustic fingerprints are introduced, the statistical characteristics of the acoustic features of the indoor space are fully utilized, the stability of the positioning system is improved, and the performance requirements of the real-time indoor positioning solution are met.