论文部分内容阅读
To enhance the GNSS localization quality degraded by signal blockage and multipath reflections in urban areas,a fingerprint positioning method of satellite signal based on probability distribution has been proposed to improve the satellite positioning accuracy.We utilize the difference of the probability distribu-tion of different positions to perform an accurate positioning in this paper.The probability distribution of the signal parameter at a certain position can be obtained with the 3-dimensional ray-tracing method,and then we can establish a fingerprint database by figuring out the probability distribution curves of the signal parameters at different locations.Finally,a matched filter method,which is used to match the actual received signal parameters with the fingerprint information in the database,can be used to realize the positioning task.The proposed fingerprint positioning method based on probability distribution greatly reduces the similarity of fingerprint information at different positions,so it can perform better than the traditional nearest neighbor method.The theoretical analysis and simulation results show that the proposed approach can improve the accuracy of positioning.