In signal-degraded environments such as urban canyons or mountainous areas many GPS signals

are blocked by natural or artificial obstacles or severely degraded; hence GPS-only cannot

guarantee an accurate and continuous positioning. The multi constellation approach, integrating

different GNSS systems, is a possible way to fill this gap. GLONASS, the Russian navigation

satellite system, is currently the main candidate as element of a multi constellation; it is nearly fully

operational and its inclusion guarantees an improvement of the satellite availability. Another

possible future component of integrated GNSS system is the European Galileo currently in a

validation phase with only 4 satellites in orbit GIOVE A/B experimental satellites and 2 Galileo for

the IOV phase.

In this work GPS/GLONASS systems are combined and relative single point algorithm

performance is assessed for different configurations in signal-degraded scenario such as urban

canyon. GPS/GLONASS multi-constellation use involves the addition of a further unknown to

estimate, i.e. the intersystem time scale offset, which requires the “sacrifice” of one measurement.

The intersystem offset is observed to be quasi-constant, so an aiding can be introduced to account

for its behavior. A similar approach can be adopted for altitude considering its typical variations in

urban scenario.

The considered estimation techniques are least squares and Kalman filter, commonly adopted to

calculate the navigation unknowns from pseudorange measurements. The least squares method uses

a model relating measurements and state with the drawback of solution unavailability during GNSS

outages (very frequent in urban areas); to improve the continuity constrained least squares

adjustments are considered. Kalman filter uses, in addition to a measurement model, a process

model expressing the unknown dynamics and allowing the state estimation in case of GNSS outage.

The main purpose of this work is the performance assessment of a multi-constellation system

relative to GPS-only adopting least squares or Kalman filter estimators.

JF - ENC 2012
ER -