Abstract: To address the problem of low accuracy caused by the non line of sight (NLOS) propagation of signals in the indoor positioning process of wireless sensor networks, a three-dimensional indoor positioning model with NLOS error elimination based on the improved Gauss Newton method was proposed. Firstly, the initial solution of the target position is obtained by using the least square algorithm based on the Euclidean distance. In order to further reduce the influence of NLOS error, the nonlinear least square estimation value is iterated according to the improved Gauss-Newton method, and the final accurate position is obtained by convergence. The experimental results show that the positioning error of the model in three-dimensional space is about 0.64 m, the maximum positioning error is not more than 1.29 m, and the probability of error less than 1.2 m is 96.5%, which is better than other positioning methods.