Résumé:
Global Navigation Satellite Systems (GNSS) have
become an integral part of all applications where mobility plays
an important role. However, The performances of GNSS-based
positioning systems can be affected in constrained environments
(urban and indoor environments), due to masking of satellites by
buildings and multipath effects.
In this paper, a comparative investigation on classical GNSS
localization algorithms in urban areas is presented and analyzed
in terms of mean squared error.
As a result, Kalman filter estimation shows the best error
performance in good environments (all satellites are in direct
sight). Nevertheless, in constrained environments, the kalman
filter and least square method show important positioning errors
because their measurement noise model is unsuitable
