SINGLE-SNAPSHOT SIGNAL SUBSPACE METHODS FOR ACTIVE TARGET LOCATION: PART Ⅰ: MULTIPLE SCATTERING CASE

摘要

This two-part paper investigates two signal subspace methods for the localization of a number of scatterers (targets) whose dimensions are small relative to the probing wavelength from active antenna array data. They are the time-reversal imaging with multiple signal classification (MUSIC) approach, and a new signal subspace method that is based on high-dimensional parameter search (a high-dimensional MUSIC). In this, first paper of the two-part sequence, a general theory of time-reversal imaging with MUSIC is developed that is applicable for the localization of arbitrary multiply scattering point targets in the near or far field via generally non-coincident arrays formed by non-collocated transmitters and receivers. The Foldy-Lax multiple scattering model is used to show that the performance of the time-reversal approach for target localization remains unchanged under Born-approximable versus non-Born-approximable (exact, multiple scattering) conditions. The general theory is illustrated with a numerical example.