First of all, a brief description of the definition, characteristics as well as the advantages of IR-UWB is presented. Besides, it describes fundamentals of several wireless positioning methods, including TOA, TDOA, AOA, etc. Also, the accuracy evaluation index and the cause of the position precision effect are presented.Secondly, this paper puts forward a Chan algorithm based on a 3D model of UWB signal positioning. IEEE 802.15.4a channel of UWB model joined the small scale fading model and extend the distance, making it practically significant to work on UWB positioning technology. In this paper, the performance of the mentioned positioning algorithm in 2-10GHz indoor office environment in 4a channel is analyzed. With the purpose of further improvement of the positioning accuracy, we filtering the TOA value of Chan algorithm by moving average algorithm.Then, this paper proposes two positioning algorithm, namely, Minimum Error Positioning and Biased Kalman Filter Positioning algorithm, to reduce the effect on positioning accuracy caused by the error of NLOS. This paper introduces the fundamentals of Minimum Error Positioning algorithm, and the Cramer-Rao lower bound (CRLB) theoretical value of the range of measurements in TDOA according with Gauss distribution. In addition, this paper performs MATLAB simulation of Minimum Error Positioning algorithm in the environment of same quantities of reference node as well as same propagation conditions. The simulation results show that no matter in line-of-sight (LOS) or non-line-of-sight (NLOS) cases, the positioning accuracy of the Minimum Error Positioning algorithm is greatly improved compared with Chan algorithm. In order to conduct positioning estimation in non-line of sight (NLOS) propagation case, a biased Kalman Filter positioning model is built. Simulation results show that the positioning accuracy is improved obviously with this Biased Kalman filtering.Next, this paper analyzes the limitations of single positioning technology. Knowing that Kalman Filter has an excellent tracking performance, the paper come up with a hybrid positioning algorithm of TDOA/AOA combined with the characteristics of AOA algorithm, along with the performance motion trail of this algorithm as well as the theoretical boundary. Simulation of the performance of this new algorithm presents that it provides better performance with this hybrid algorithm as long as the AOA measurement in service base station is precisely enough. A 3D positioning system is built in laboratory to verify this hybrid algorithm, receiving a result that the positioning accuracy is Cm level with an accuracy of 1 centimeter in the best situation.At last, this paper provides a single base station UWB positioning system with high accuracy based on the antenna array. In this system, the amplitude of the received pulse signal on the antenna array of the reference base station is measured instead of previously measuring the tiny time differences. Combined with the beam pattern of the antennas, it is easy to precisely calculate the arrival angle of the signal.Thus, it is convenient to safely obtain the accurate location of the positioning label. In experiment, some improvements of the ultra-wideband pulse generator are taken,resolving the severe trailing problem of the output signals. In the end, function is set up to dispel the influence that the attenuation of the signal amplitude has on the positioning accuracy. In this way, we finally obtain a TOA positioning accuracy of 2 centimeters, as well as a AOA accuracy of 1 centigrade.