Time Domain Inverse Scattering of a Two-Dimensional Homogenous Dielectric Object With Arbitrary Shape by Particle Swarm Optimization

Abstract

Abstract-This paper presents a computational approach to the twodimensional time domain inverse scattering problem of a dielectric cylinder based on the finite difference time domain (FDTD) method and the particle swarm optimization (PSO) to determine the shape, location and permittivity of a dielectric cylinder. A pulse is incident upon a homogeneous dielectric cylinder with unknown shape and dielectric constant in free space and the scattered field is recorded outside. By using the scattered field, the shape and permittivity of the dielectric cylinder are reconstructed. The subgridding technique is implemented in the FDTD code for modeling the shape of the cylinder more closely. In order to describe an unknown cylinder with arbitrary shape more effectively, the shape function is expanded by closed cubicspline function instead of frequently used trigonometric series. The inverse problem is resolved by an optimization approach, and the global searching scheme PSO is then employed to search the parameter space. Numerical results demonstrate that, even when the initial guess is far away from the exact one, good reconstruction can be obtained. In addition, the effects of Gaussian noise on the reconstruction results are investigated. Numerical results show that even the measured scattered E fields are contaminated with some Gaussian noise, PSO can still yield good reconstructed quality.

Time-Domain Inverse Scattering of a Two-Dimensional Metallic Cylinder in Slab Medium Using Asynchronous Particle Swarm Optimization

“…By combining the FDTD method and the APSO, good reconstructed results are obtained. The key differences between PSO [11] and APSO are about the convergence speed, the computation time and the accuracy, since APSO includes "damping boundary condition" scheme and mutation scheme. The inverse problem is reformulated into an optimization one, and then the global searching scheme APSO is employed to search the parameter space.…”

“…Moreover, multiple frequencies can be investigated without any extra computational effort. Therefore, various time domain inversion approaches are proposed [6][7][8][9][10][11] that could be briefly classified as the neural networks [6], the iterative approach: Born iterative method (BIM) [7], and gradientbased method [8], and optimization approach [9][10][11]. Traditional iterative inverse algorithms are founded on a functional minimization via some gradient-type scheme.…”

“…In the past twenty years, the inversion techniques are developed intensively for the microwave imaging both in frequency domain and time domain [3][4][5][6][7][8][9][10][11][12][13][14][15][16]. Most of the inversion techniques are investigated for the inverse problem using only single frequency scattering data (monochromatic source) [3][4][5].…”

Inversion of Lossy Dielectric Profiles Using Particle Swarm Optimization

“…On the other hand, global optimization techniques do not require priori information about the model, for this reason convergence is reached after relatively large number of iterations. The most recent global techniques which are used in electromagnetic inversion problems are the genetic algorithms (GA) [10,11] and the particle swarm optimization techniques (PSO) [12][13][14][15]. The swarm technique has proven to outperform GA due to many reasons.…”

Microwave Imaging Within the Interval Analysis Framework

“…In order to avoid nonuniqueness and instability as well as to prevent the retrieval of false solutions [28], several inversion strategies have been proposed based on (a) a suitable definition of the integral equations either in exact [29,30] or approximated [31][32][33][34][35] forms to model the scattering phenomena, (b) the exploitation of the available a-priori information on some features of the scenario/scatterers under test [15,[36][37][38][39] or/and the knowledge of input-output samples of data and reference solutions [40][41][42] and/or the information acquired during the inversion process [43][44][45][46][47], and (c) the use of suitable global optimization strategies [48][49][50][51][52][53][54][55]. Whatever the approach, inversion methods generally consider an optimization step aimed at minimizing/maximizing a suitably defined data-mismatch cost function through gradient or evolutionarybased algorithms with still not fully resolved drawbacks.…”