With more and more attention paid to the service based on location information, wireless location technology has also been developed rapidly. Among them, ultra-wideband (UWB) technology has the characteristics of high measurement accuracy and low power consumption, which is more and more widely used in wireless positioning system. However, the transmission of wireless signals is easy to be interfered with by the environment, which leads to a sharp decline in the positioning accuracy under signal interference, which restricts the application and promotion of this positioning technology. In this paper, a nonlinear fitting, spatial error compensation, measurement error compensation, and machine learning-based pattern recognition model are established by using accurate spatial positioning equations to solve the problem of UWB accurate positioning under signal interference.