Inverse scattering with fixed - energy data ∗

The Newton-Sabatier method for solving inverse scattering problem with fixed-energy phase shifts for a sperically symmetric potential is discussed. It is shown that this method is fundamentally wrong in the sense that its foundations are wrong: in general it cannot be carried through because its basic integral equation may be not solvable for some r > 0 and in this case the method breaks down; the basic ansatz of R.Newton is incorrect: in general, the transformation kernel does not have the form postulated in this ansatz; consistency of the method is not established, and some of the physical conclusions, e.g., existence of the transparent potentials, are not proved. A mathematically justified method for solving the three-dimensional inverse scattering problem with fixed-energy data is described. This method is developed by A.G.Ramm for exact data and for noisy discrete data, and error estimates for this method are obtained. Difficulties of the numerical implementation of the inversion method based on the Dirichlet-to-Neumann map are pointed out and compared with the difficulty of the implementation of the Ramm's inversion method.