A new method for multipath interference suppression in through-the-wall UWB radar imaging

When through-the-wall ultra-wideband (UWB) synthetic aperture radar (SAR) is applied to imaging, the interference of multipath spreading between targets will introduce ghosts into the SAR image. This paper proposes a novel technique for through-the-wall UWB SAR imaging with the capability to suppress multipath ghosts. A multipath scattering model is first developed for through-the-wall UWB SAR. By analyzing the model, we obtain an important feature of multipath ghosts in through-the-wall UWB SAR images: aspect dependence. Based on the ghost's aspect dependence, we develop a hidden Markov model (HMM)-based method for suppression the ghosts in the SAR images. Simulations show that the proposed method can improve the peak value ratio of real targets and ghosts by about 15 dB.

I. Introduction

Conventional through-the-wall SAR image formation assumes that the physical mechanism of a given scattering event involves only a single reflection of the electromagnetic wave with a portion of the extended objectof- interest [1]. But in practice, it is not so. The result of the multipath scattering between targets is the formation of ghosts in the SAR image, which will seriously deteriorate the detection and recognition performance.