Mobile localization using subspace tracking

Abstract

Time-of-arrival (TOA) and direction-of-arrival (DOA) based methods are essential for mobile positioning in UMTS networks. Because of the presence of closely spaced multipaths, timing errors will occur when standard correlation based estimation methods are applied. To mitigate this problem, we propose here to use first a rough estimate of the desired TOA followed by T-MUSIC algorithm for a refinement of this estimate. The latter algorithm is implemented using a modified version of the adaptive subspace tracking algorithm OPAST (Orthogonal Projection Approximation Subspace Tracking) for the extraction and tracking of the principal subspace of a positive Hermitian covariance matrix. Computer simulations prove the numerical stability of the modified OPAST and the good estimation & tracking performance of the proposed method in typical propagation environments. © 2005 IEEE.