A NOTE ON MOST FAVORABLE ARRAY GEOMETRIES FOR DOA ESTIMATION AND ARRAY INTERPOLATION

Given an n-element linear array with the fixed positions x(1) and x(n) of the leftmost and rightmost array sensors, it is shown that the sto chastic Cramer-Rao bound (CRB) and MUSIC performance depend on positio ns of the remaining n-2 sensors within the interval [x(1), x(n)]. The asymptotic performance of interpolated array approach shows similar de pendence. The most favorable geometries are unrealizable for q < n-1 b ecause the array sensors tend to form q+1 point clusters, where q is t he number of sources. An interesting consequence of these facts is tha t for certain realizable nonuniform linear array (NULA) geometries, in terpolated root-MUSIC with virtual uniform linear array (ULA) of the l ength x(n)-x(1) has better asymptotic performance than conventional ro ot-MUSIC applied to the real ULA of the same length.