Subspace-based optimization method coupled with multiplicative regularization for edge-preserving inversion

We present a multiplicative-regularized subspace-based optimization method for solving inverse scattering problems where scatterers are piecewise homogeneous. The proposed inversion algorithm not only has the advantages of low computational complexity and better stability against noise, but also exhibits the edge-preserving ability. The extension of the method to the twofold subspace-based optimization method is discussed. Numerical simulations validate the proposed algorithm.