Near-field localization algorithm based on sparse reconstruction of the fractional lower order correlation vector

Abstract

This paper addresses the issue of joint direction-of-arrival (DOA) and range estimation of near-field signal under impulsive noise environments modeled by a-stable distribution. Since a-stable distribution does not have finite second-order statistics, the DOA and range estimation problem under impulsive noise environment can be decoupled in the fractional lower order correlation domain. Then, the two dimensional positioning problem is transformed into two one dimensional parameter estimation problems which can be solved by the sparse reconstruction of the fractional lower order correlation vector. The computer simulation results demonstrate that the proposed algorithm outperform the second order correlation-based methods.