Bayesian Pursuit algorithm for sparse representation

Abstract

This paper addresses the problem of scene reconstruction, incorporating wall-clutter mitigation, for compressed multi-view through-the-wall radar imaging. We consider the problem where the scene is sensed using different reduced sets of frequencies at different antennas. A joint Bayesian sparse recovery framework is first employed to estimate the antenna signal coefficients simultaneously, by exploiting the sparsity and correlations between antenna signals. Following joint signal coefficient estimation, a subspace projection technique is applied to segregate the target coefficients from the wall contributions. Furthermore, a multitask linear model is developed to relate the target coefficients to the scene, and a composite scene image is reconstructed by a joint Bayesian sparse framework, taking into account the interview dependencies. Experimental results show that the proposed approach improves reconstruction accuracy and produces a composite scene image in which the targets are enhanced and the background clutter is attenuated. ABSTRACT This paper addresses the problem of scene reconstruction, incorporating wall-clutter mitigation, for compressed multi-view through-the-wall radar imaging. We consider the problem where the scene is sensed using different reduced sets of frequencies at different antennas. A joint Bayesian sparse recovery framework is first employed to estimate the antenna signal coefficients simultaneously, by exploiting the sparsity and correlations between antenna signals. Following joint signal coefficient estimation, a subspace projection technique is applied to segregate the target coefficients from the wall contributions. Furthermore, a multitask linear model is developed to relate the target coefficients to the scene, and a composite scene image is reconstructed by a joint Bayesian sparse framework, taking into account the interview dependencies. Experimental results show that the proposed approach improves reconstruction accuracy and produces a composite scene image in which the targets are enhanced and the background clutter is attenuated. Index Terms-Multi-view through-the-wall radar imaging, wall clutter mitigation, compressed sensing, joint Bayesian sparse recovery.

Boltzmann Machine and Mean-Field Approximation for Structured Sparse Decompositions

“…2) BG model: As mentioned in the introduction, BG model (6)- (7) has already been considered in some contributions ( [28], [29], [32], [33]) and under the marginal formulation (3) in [35], [36]. However all these contributions differ from the proposed approach by the estimation problem and the practical procedure introduced to solve it.…”

“…Another model on z based on BG variables is as follows: the elements of the sparse vector are defined as the multiplication of Gaussian and Bernoulli variables. This model has been exploited in the contributions [28], [29], [32], [33] and will be considered in the present paper. These two distinct hierarchical BG models share a similar marginal March 15, 2012 DRAFT expression of the form:…”

“…March 15, 2012 DRAFT Thus, in [28], [29], [32], [33], the authors focus on the joint MAP estimation problem (17). They then propose different greedy procedures to solve it, some of them are explicitly related to standard deterministic algorithms, as BMP, BOMP [32] or SBR [33] and their respective standard homologues MP, OMP and OLS.…”

An iterative bayesian algorithm for block-sparse signal reconstruction

“…To model the block-sparse sources w, we introduce two hidden random processes, s and θ [14]- [16]. The binary vector s ∈ {0, 1} M describes the support of w, denoted S, while the vector θ ∈ R M represents the amplitudes of the active elements of w. Hence, each element of the source vector w can be characterized as…”

Compressed Sensing ISAR Reconstruction Considering Highly Maneuvering Motion

“…The solutions of the optimization problem involved in CS can be obtained by several techniques, such as orthogonal matching pursuit (OMP) [5], Bayesian CS [6], Bayesian matching pursuit [7,8], modified quasi-Newton method [9], etc. Similarly, applications of CS have been demonstrated in different works, such as single-pixel remote sensing [10], tomographic SAR [11] , through-the-wall imaging for stationary and moving targets [12] , radar imaging [13][14][15], passive radar imaging [16], sparse microwave imaging of perfect electric conducting targets [17], etc.…”