Analytical results towards fuzzy error bound of indoor neighbor matching based positioning algorithm

Abstract

In this paper, we derive out an analytical result towards fuzzy error bound of indoor Neighbor Matching based Positioning Algorithm (NMPA). First of all, in a typical Line-of-sight (LOS) environment, we utilize the fuzzy comprehensive evaluation approach to obtain the fuzzy membership of the target in fingerprint matching. Second, we present an analysis of the theoretical relationship between the positioning error of NMPA and different environmental parameters. Third, we work out the fuzzy closed form solution to the positioning error of NMPA concerning the size of environment, the spacing of Reference Points (RPs), number of neighbors, and positions of Access Points (APs). Finally, relevant experiments verify that the fuzzy error bound proposed in this paper can reflect the influence of different factors of the environment on the performance of the positioning system, thereby, the positioning accuracy can be improved and the deployment costs can be reduced by optimizing the environmental parameters.