Enhanced through-the-wall radar imaging using Bayesian compressive sensing

Tang, Van Ha; Bouzerdoum, A.; Phung, Son Lam; Tivive, Fok Hing Chi

doi:10.1117/12.2014814

Cited by 11 publications

(8 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In TWRI, this has the advantage of reducing the number of measurement samples and data acquisition and processing time. A number of CS-based methods were proposed for TWRI in recent years [5,8,23,24,30,31]. In this section, we review briefly the CS-based approach for solving the image formation problem as an inverse problem using the single measurement vector model.…”

Section: Single-polarization Imaging Using Smv Modelmentioning

confidence: 99%

See 1 more Smart Citation

Sparsity-Aware Human Motion Indication

Bouzerdoum

Yang

Tivive

2017

Compressive Sensing for Urban Radar

View full text Add to dashboard Cite

Discrimination of targets can be improved significantly by analyzing the polarization of scattered electromagnetic waves. In radar imaging, the target image can be enhanced by combining measurements from different polarizations. In this chapter, we propose a joint image formation and fusion approach for multipolarization through-the-wall radar imaging, using compressive sensing (CS). The measurements from different polarization channels are processed jointly using the multiple measurement vector (MMV) model to produce several images of the scene, each corresponding to a polarization channel. Furthermore, the measurement vectors are fused together to form a composite measurement vector, which yields a composite image of the scene. The advantage of fusing the measurement vectors before image formation is that the measurement noise is reduced and the target information is enhanced, which leads to a more informative composite image. The MMV model enforces the same sparsity support for all formed images by reinforcing target information across channels and attenuating noise. Experimental results are presented using simulated and real data. Analysis and comparison of experimental results demonstrate the effectiveness of the proposed through-the-wall radar imaging approach, especially in the presence of high-measurement noise.

show abstract

Section: Single-polarization Imaging Using Smv Modelmentioning

confidence: 99%

“…Recently, CS has been considered for radar imaging due to its ability to reconstruct a high-resolution image from a reduced set of measurements [2,5,17,18,23,24,31]. The scene reconstruction is posed as an inverse problem, whereby a spatial map of reflections is formed from radar measurements.…”

Section: Introductionmentioning

confidence: 99%

Sparsity-Aware Human Motion Indication

Bouzerdoum

Yang

Tivive

2017

Compressive Sensing for Urban Radar

View full text Add to dashboard Cite

show abstract

“…Hence, the signal correlations across the antennas are exploited through the estimation of the hyper-parameters. The reader is referred to [19,20] for a detailed description of the algorithm. Once the hyper-parameters α are estimated, the wavelet coefficients θ m are obtained by the mean of the posterior given in Eq.…”

Section: Joint Bayesian Sparse Signal Modelmentioning

confidence: 99%

Enhanced wall clutter mitigation for compressed through-the-wall radar imaging using joint Bayesian sparse signal recovery

Tang

Bouzerdoum

Phung

et al. 2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

This paper addresses the problem of wall clutter mitigation in compressed sensing through-the-wall radar imaging, where a different set of frequencies is sensed at different antenna locations. A joint Bayesian sparse approximation framework is first employed to reconstruct all the signals simultaneously by exploiting signal sparsity and correlations between antenna signals. This is in contrast to previous approaches where the signal at each antenna location is reconstructed independently. Furthermore, to promote sparsity and improve signal reconstruction accuracy, a sparsifying wavelet dictionary is employed in the sparse signal recovery. Following signal reconstruction, a subspace projection technique is applied to remove wall clutter, prior to image formation. Experimental results on real data show that the proposed approach produces significantly higher reconstruction accuracy and requires far fewer measurements compared to the single-signal model, where each antenna signal is reconstructed independently.

show abstract

“…In the literature, different CS algorithms were used for radar‐image reconstruction [1, 3, 13–23]. Basis pursuit (BS), Lasso, and other convex relaxation algorithms that are based on

ℓ_{1}

‐minimisation were used in [13, 21–23].…”

Section: Introductionmentioning

confidence: 99%

“…Bayesian CS usually outperforms other CS algorithms [27, 28]. The authors in [20, 29–34] proposed Bayesian CS framework for radar‐imaging applications. In [20, 29], a hierarchical‐prior model was invoked to represent the priori probability distribution.…”

Section: Introductionmentioning

confidence: 99%