The paper is aimed at the demand for the improvement of the aircraft approach guidance and positioning ability. In response to the problems of the pseudolites measure fluctuations and a large amount of positioning calculation, an improved Unscented Kalman Filter (UKF) algorithm based on the hypersphere simplex sampling strategy is proposed. Firstly, an adaptive factor is introduced to adjust the weight ratio of the state prediction information and the measurement information in the filtering solution. While reducing the number of Sigma sampling points, the filtering stability is improved. In addition, the time-varying measurement noise is estimated through the method of online real-time adjustment by the Fuzzy Inference System (FIS), which enhances the positioning accuracy. The simulation verifies the convergence and effectiveness of the filtering algorithm. The proposed algorithm in the paper can provide high-precision position service for aircraft approach process.