The invention discloses a bluetooth indoor micro-positioning system based on an ELMAN neural network. The system comprises a system host, a bluetooth anchoring point and a positioning target, the system host comprises an embedded processor, a bluetooth main node and a communication module, the bluetooth main node (main node bluetooth module) adopts a low-power-consumption bluetooth wireless data transmission module RL-CC2541-S3, a SIM900A is used for communication with a host computer, a CC2541 acquires a host signal intensity indication required by positioning, one bluetooth module is fixed on the positioning target, the system host and the bluetooth anchoring point are arranged in a positioned indoor area, the bluetooth module fixed on the positioning target operates in a host mode, andthe bluetooth main node and the bluetooth anchoring point operate in a slave mode. According to the system, the precision and the stability of the positioning technology are improved, certain adaptability to the environment is achieved, and technical support is provided for implementation of a complex positioning algorithm.