Accuracy Assessment of a High Sensitivity GPS Based Pedestrian Navigation System Aided by Low-Cost Sensors

Key words: pedestrian navigation, high sensitivity GPS Using long total dwell times, acquisition and weak GPS signal tracking in degraded signal environments are possible. A technology utilizing such long integration methods is called High Sensitivity GPS (HSGPS). A previous hardware in-the-loop GPS signal simulator test has demonstrated that, for a stand-alone HSGPS implementation provided by SiRF Technologies Inc., signals as weak as-186 dBW could be tracked by the receiver (MacGougan et al., 2002). This corresponds to 25-30 dB weaker signals than the typical outdoor line-of-sight GPS signals. However, measurement noise increases as signal power decreases and the receiver tracking loops become susceptible to possible tracking of cross-correlation or echo-only signals that can lead to severe position and velocity estimation errors. This paper presents the results of the accuracy analysis of the sensor aided HSGPS receiver for pedestrian navigation in signal-degraded environments. Description of the downtown test and the equipment used is given. Analysis of the stand-alone HSGPS results with Receiver Autonomous Integrity Monitoring (RAIM) techniques is presented. Brief analysis of gyro bias estimation problems with GPS is presented. Kalman filter integration methodology is presented and the results are analyzed. Introduction Recently, significant research results have taken place in the area of pedestrian navigation. Most of the work so far has concentrated on the design, implementation and integration aspects of inertial sensors with conventional GPS. In most indoor environments and urban canyons, conventional GPS barely provides any position/velocity fixes due to poor observation availability. Thus, such systems mostly navigate in a pure inertial or Pedestrian Dead Reckoning (PDR) mode (e.g., Levi et al., 1999) once a GPS solution is unavailable. If an HSGPS receiver is used as part of the system, availability improves but new challenges arise. The availability of the HSGPS solution becomes truly remarkable considering the environment, but poor statistical estimates of the measurement errors present a challenge for a conventional Kalman filter (MacGougan et al., 2003).