For the problems such as high ranging errors of ultra wide band(UWB) in the non-line-of-sight(NLOS) environment, a NLOS identification method is proposed based on the position difference between the first path and strongest path, the power difference between the received signal and the first path, and the distance residuals. This algorithm is easy to calculate and has high identification accuracy. Further, a NLOS error mitigation method with received signal strength and time of arrival fusion is proposed based on the biased Kalman filtering and maximum likelihood estimation algorithms. Finally, experiments in dynamic and static scenarios are carried out to validate the proposed algorithm. The experimental results show that the average identification accuracy of the proposed method reaches 95.42% for NLOS scenes, and the average distance error of positioning is within 0.14m.