The identification of a spherically symmetric potential by its phase shifts
is an important physical problem. Recent theoretical results assure that s
uch a potential is uniquely defined by a sufficiently large subset of its p
hase shifts at any one fixed energy level. However, two different potential
s can produce almost identical phase shifts. To resolve this difficulty we
suggest the use of phase shifts corresponding to several energy levels. The
identification is done by a nonlinear minimization of the appropriate obje
ctive function. It is based on a combination of probabilistic global and de
terministic local minimization methods. The Multilevel Single-Linkage Metho
d (MSLM) is used for the global minimization. A specially designed Local Mi
nimization Method (LMM) with a Reduction Procedure is used for the local se
arches. Numerical results show the effectiveness of this procedure for pote
ntials composed of a small number of spherical layers.