Weighting and Mitigation of Multipath Effects in GPS Range Measurements in an Urban Environment for Pedestrian Navigation

摘要

In this paper we will show how to process rangernmeasurements in a high sensitive GPS receiver in anrnurban environment by mitigating multipath effects in arnreal world scenario. We will compare a c0 weighting ofrnrange measurements approach to a maximum likelihoodrnapproach where multipath effects are identified andrnmitigated. A previous approach of a maximum likelihoodrnestimator is evaluated and enhanced for use in arnpedestrian navigation system. These enhancementsrninclude a higher robustness in severe multipathrnenvironments and a gain in position accuracy by applyingrnweightings to the range measurements corrected by thernmaximum likelihood estimator in respect of satelliterngeometry and c0. Such a high sensitive GPS receiver isrnadapted to our Dual-IMU pedestrian navigation systemrnwhere a tight integration of an Inertial Measurement Unitrn(IMU), a barometric sensor, a magnetic compass and thernGPS sensor is implemented. The Dual-IMU comprises arnfoot mounted and a torso mounted IMU. FurthermorernZero Velocity Updates (ZVUs) during the stance phase ofrnthe foot mounted IMU reduce the drift of the MEMSrnbased inertial system. For the Tightly Coupled GPS/INSrndata fusion the obtained GPS range measurements have tornbe weighted properly in the measurement step of thernKalman filter.