Android智能手机双频GNSS伪距差分动态定位性能分析

Performance Analysis of Dual-frequency GNSS Pseudo-range Differential Dynamic Positioning for Android Smartphones

关键词:  Android智能手机  双频GNSS观测值  伪距差分改正数  定位性能分析

Key words:  Android smartphone  Dual-frequency GNSS observation  DGNSS correction  Positioning performance analysis

摘要:

随着智能手机全球导航卫星系统（GNSS）天线和芯片从单频单模向多频多模快速发展，基于其所衍生出来的位置服务（LBS）应用极大地便利了大众用户的日常生活。然而受限于低成本、低功耗的信号接收与处理单元，手机在无增强信息的情况下仅依赖伪距单点定位难以为用户提供稳定、高精度的导航服务。因此，基于小米8手机（Mi8）的GNSS双频原始数据，采用非组合的双频伪距观测值、载波历元差分观测值和多普勒观测值构建了滤波定位模型，并引入伪距差分数据，以提升手机定位的连续性和精度。在较复杂环境下开展了行人和车载实验，实测结果表明：双频定位精度与单频相比提升了15%~30%，伪距差分定位精度和单点定位相比提升了5%~20%，行人和车载双频伪距差分定位的平面位置误差分别为0.65m和1.03m，基本满足手机用户在城市复杂环境下的高精度定位服务需求。

Abstract:

As GNSS (Global Navigation Satellite System) antennas and chips in smartphones develop rapidly from single-frequency and single-system to multi-frequency and multi-system, mass users benefit greatly from the application of location based services (LBS) supported by smartp-hones. Without any enhanced information, using standard point positioning (SPP) only is difficult to provide mass users with reliable and high-precision navigation services, which is limited by low-cost and low-power hardware of smartphones. A filter model is built using non-combined dual-frequency pseudo-range, time-differenced carrier phase and Doppler from Xiaomi 8, and DGNSS corrections are introduced in order to improve the positioning continuity and accuracy. The results of the pedestrian and vehicle experiments show that the accuracy of dual-frequency positioning is increased by 15%~30% compared with single-frequency positioning, and the accuracy of DGNSS positioning is increased by 5%~20% compared with SPP. The horizontal RMS position errors of the pedestrian and vehicle mode are 0.65m and 1.03m, respectively, which can generally meet the needs of mass users for high-precision navigation services in urban environments.