Inversion of phaseless total field data by reconstructing the equivalent radiating currents

Xiang, Yin; Li, Lianlin; Li, Fang

doi:10.1109/apmc.2008.4958524

Cited by 1 publication

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is along the line of our previous work [21] but gives more contents. Though many other excellent algorithms of phase retrieval have been developed, how to avoid the local solution by incorporating suitable prior information such as support and zero-point information of unknown signals is still a challenging problem [17].…”

Section: Introductionmentioning

confidence: 82%

A Novel Phase Retrieval Approach for Electromagnetic Inverse Scattering Problem With Intensity-Only Data

Xiang¹,

Li²,

Li³

2010

PIER M

View full text Add to dashboard Cite

Abstract-To measure the phase of signal with very high working frequency such as THz, and optics band is still a challenging problem. In this paper, based on the relationship between radiating current and measured intensity of electrical field a novel phase retrieval algorithm has been developed. As opposed to the existing approaches of phase retrieval where usually the Fourier coefficients of measured data will be firstly reconstructed, the proposed approach is to reconstruct the so-called radiating currents, with more physical meaning than the former. It has a much smaller number of freedoms of radiating current than that of measurements, which means that the obtained equations are over-determined. Thus one can efficiently model the intensity of measured electric field via the radiating part, and reconstruct it quickly and stably. The novelty is that this physical consideration 1) leads to efficiently avoiding false solutions due to the ill-posedness of phase retrieval problem, and 2) offers a good initial guess for inverse scattering based imaging algorithm. Importantly, a closed-form formulation of phase retrieve also has been derived when the intensity of incident wave is much stronger than one of the scattered wave, for example, for the weak scattering objects. Finally, several numerical experiments are provided to show the high performance of proposed algorithm.

show abstract

Section: Introductionmentioning

confidence: 82%