The invention belongs to the technical field of wireless communication, and discloses an RSSI indoor positioning and distance measuring method based on neural network learning, and an indoor positioning platform, wherein communication is established between a target node and an anchor node, and collected data is stored in a set RSSI[i] = {RSSIi1, RSSIi2, ..., RSSIiN}; setting a screening probability p, determining an upper limit value RSSImax and a lower limit value RSSImin according to a Gaussian model of the RSSI; storing the RSSI in a set RSSI[i] in a range of [RSSImin, RSSImax] into a setRSSI_gauss[i]; averaging RSSI values in the set RSSI_gauss[i]; training a strong separator by combining an algorithm of a particle swarm optimization neural network and an idea of an iterator; converting the RSSI into a distance between the anchor node and the target node by using the strong separator; and obtaining a solution of the target node by using a maximum likelihood estimation method. According to the method, the workload is reduced, the larger error existing in a single algorithm is reduced, and the accuracy of converting the RSSI value into the distance is improved, so that the positioning accuracy is improved.