Detection of buried object using RF tomography under high-order born approximation assumption

Abstract EN

Radiofrequency tomography (RFT) is frequently employed for the detection and reconstruction of images of deeply buried targets and underground facilities.

This research combines ground penetrating radar with RFT for assessment of the environment, pinpointing the location of a target, and creating a target image of suc buried objects as landmines or unexploded ordnance.

Following data collection from a number of transmitters and receivers, the inverse scattering principle is employed for the computation of dielectric properties.

For the resolution of the nonlinear inverse scattering difficulty, a high-order Born approximation is employed.

The transmitters beam appropriate low-frequency narrowband waveforms at the earth’s surface; the wave front thus created encounters underground objects, which scatters electromagnetic energy in every direction.

Receivers take a sample of the scattered signal, retrieving their phases, and this information is relayed to a central processor.

Once the captured data has been put through adaptive processing, an image of the underground object can be constructed.

This paper proposes the reduction in computational complexities and an improved image reconstruction through a simple and efficient form of higher-order Born approximation.

The proposed method is based on updating the reconstructing domain at each Born approximation process.

This allows high-quality resolution imagery to be constructed when working with significant contrast objects simply through the resolution of a linear inverse-scattering problem.

A Conjugate Gradient algorithm has been employed for the linear inversion.

Numerical testing has demonstrated that this suggested methodology is both valid and efficient