Abstract: For the mobile terminal positioning in the cellular networks, the error of non-line-of-sight(NLOS) environment is the main reason which leads to the decrease of the positioning accuracy. To reduce the influence of NLOS error, a positioning method based on signed weighted residuals was proposed in this paper. Firstly Chan algorithm was adopted to calculate the initial position of mobile station, and then the position with signed weighted residuals model was revised, and finally the position of the mobile station with the signed residual adjuvant Taylor series expansion method of iteration was obtained. The simulation results show that compared with the weighted squared residuals based method, the positioning accuracy of the signed weighted residuals based method is increased by about 14.82% and it can suppress the influence of NLOS more effectively.