Ultra-wide-band technology has important application value for location services due to its high ranging accuracy and penetration performance. In the actual high-density positioning environment, the traditional positioning algorithm is affected by non-line-of-sight error and multipath effect, and it is difficult to accurately calculate the actual position coordinates in real time. Although increasing the number of base stations can effectively improve the accuracy of positioning, its cost also increases. Aiming at the improvement of the accuracy and robustness of positioning, an ultra-wideband positioning method based on support vector machine was proposed to solve the problem of poor real-time performance and low positioning accuracy of ultra-wideband in high-density indoor positioning. A support vector machine model based on TDOA(TDOA, time difference of arrival) was given, with focus on transformation of the problem of location into the problem of classification. The support vector machine classification model was established by TDOA values and coordinate values. The one-to-one classification model was used to solve the coordinate values and improve the coordinate solution speed. The simulation results show that in the high-density real-time positioning, compared with the traditional Chan algorithm and Taylor algorithm, the method has higher real-time performance when the positioning accuracy is similar, which meets requirements for the actual positioning with its low power consumption, fast and high precision.
