Reconstruction of Isolated Objects by PSO-based Inverse Scattering Method

Inverse scattering problems are ill-posed and usually be transformed into optimization problems. Conventionally, the searching directions in these optimization processes are calculated by certain gradient-based deterministic algorithms. In the absence of a priori knowledge, there is a great risk of obtaining a local extremum result. In this work, a customized particle swarm optimization(PSO) method is utilized to search the globally optimal solutions of inverse problems. The introduction of random factors can help inversion optimization evade local extrema. A numerical example is included to demonstrate the effectiveness of the proposed method for inverse problems.