Search

In this study, we consider algorithm unfolding for the multiple measurement vector (MMV) problem in the case where only few training samples are available. Algorithm unfolding has been shown to empirically speed-up in a data-driven way the convergence of various classical iterative algorithms, but for supervised learning, it is important to achieve this with minimal training data. For this, we consider learned block iterative shrinkage thresholding algorithm (LBISTA) under different training strategies. To approach almost data-free optimization at minimal training overhead, the number of trainable parameters for algorithm unfolding has to be substantially reduced. We therefore explicitly propose a reduced-size network architecture based on the Kronecker structure imposed by the MMV observation model and present the corresponding theory in this context. To ensure proper generalization, we then extend the analytic weight approach by Liu and Chen to LBISTA and the MMV setting. Rigorous theoretical guarantees and convergence results are stated for this case. We show that the network weights can be computed by solving an explicit equation at the reduced MMV dimensions which also admits a closed-form solution. Toward more practical problems, we then considered convolutional observation models and show that the proposed architecture and the analytical weight computation can be further simplified and thus open new directions for convolutional neural networks. Finally, we evaluate the unfolded algorithms in numerical experiments and discuss connections to other sparse recovering algorithms. [ABSTRACT FROM AUTHOR]

In this study, we consider algorithm unfolding for the multiple measurement vector (MMV) problem in the case where only few training samples are available. Algorithm unfolding has been shown to empirically speed-up in a data-driven way the convergence of various classical iterative algorithms, but for supervised learning, it is important to achieve this with minimal training data. For this, we consider learned block iterative shrinkage thresholding algorithm (LBISTA) under different training strategies. To approach almost data-free optimization at minimal training overhead, the number of trainable parameters for algorithm unfolding has to be substantially reduced. We therefore explicitly propose a reduced-size network architecture based on the Kronecker structure imposed by the MMV observation model and present the corresponding theory in this context. To ensure proper generalization, we then extend the analytic weight approach by Liu and Chen to LBISTA and the MMV setting. Rigorous theoretical guarantees and convergence results are stated for this case. We show that the network weights can be computed by solving an explicit equation at the reduced MMV dimensions which also admits a closed-form solution. Toward more practical problems, we then considered convolutional observation models and show that the proposed architecture and the analytical weight computation can be further simplified and thus open new directions for convolutional neural networks. Finally, we evaluate the unfolded algorithms in numerical experiments and discuss connections to other sparse recovering algorithms.

We propose a robust approach to recovering jointly sparse signals in the presence of outliers. The robust recovery task is cast as a convex optimization problem involving a minimax concave loss function (which is weakly convex) and a strongly convex regularizer (which ensures the overall convexity). The use of the nonconvex loss makes the problem difficult to solve directly by the convex optimization methods even with the well-established firm shrinkage. We circumvent this difficulty by reformulating the problem via the Moreau decomposition so that the objective function becomes a sum of convex functions that can be minimized by the primal-dual splitting method. The parameter designs/ranges for the present specific case are derived to ensure the convergence. We demonstrate the remarkable robustness of the proposed approach against outliers by extensive simulations to the application of multi-lead electrocardiogram as well as synthetic data. [ABSTRACT FROM AUTHOR]

Massive MIMO is a variant of multiuser MIMO, in which the number of antennasM at the base-station is very large. It has been observed that in many realistic propagation scenarios, although the user channel vectors have a very high-dim M, they lie on low-dim subspaces because of their limited angular spread (spatial correlation). This low-dim subspace structure remains stable across several coherence blocks and can be exploited to improve the system performance. In a recent work, we addressed this problem and proposed a very effective novel algorithm referred to as Approximate Maximum-Likelihood (AML), which was formulated as a semi-definite program (SDP). In this paper, we address two problems left open in our previous work, namely, computational complexity and tracking. We propose a new algorithm that is reminiscent of Multiple Measurement Vectors (MMV) problem in Compressed Sensing and prove that it is equivalent to the AML Algorithm for sufficiently dense angular grids. It has also a very low computational complexity and is able to track the sharp transitions in the channel statistics very quickly. We provide numerical simulations to assess the estimation/tracking performance of our proposed algorithm, with a particular emphasis on situations where a direct implementation of the SDP is infeasible in real-time. Our proposed algorithm is of independent interest in applications other than massive MIMO. We provide numerical simulations to compare the performance of our algorithm with that of other related subspace estimation algorithms in the literature. [ABSTRACT FROM AUTHOR]

The standard sparse representation aims to reconstruct sparse signal from single measurement vector which is known as SMV model. In some applications, the SMV model extend to the multiple measurement vector (MMV) model, in which the signal consists of a set of jointly sparse vectors. In this paper, efficient algorithms based on split Bregman iteration are proposed to solve the MMV problems with both constrained form and unconstrained form. The convergence of the proposed algorithms is also discussed. Moreover, the proposed algorithms are used in magnetic resonance imaging reconstruction. Numerical results show the effectiveness of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Abstract: The problem of recovering a sparse solution from Multiple Measurement Vectors (MMVs) is a fundamental issue in the field of signal processing. However, the performance of existing recovery algorithms is far from satisfactory in terms of maximum recoverable sparsity level and minimum number of measurements required. In this paper, we present a high-performance recovery method which mainly has two parts: a versatile recovery framework named RPMB and a high-performance algorithm for it. Specifically, the RPMB framework improves the recovery performance by randomly projecting MMV onto a subspace with lower and tunable dimension in an iterative procedure. RPMB provides a generalized framework in which the popular ReMBo (Reduce MMV and Boost) algorithm can be regarded as a special case. Furthermore, an effective algorithm that can be embedded in RPMB is also proposed based on a new support identification strategy. Numerical experiments demonstrate that the proposed method outperforms state-of-the-art methods in terms of recovery performance. [Copyright &y& Elsevier]

The multiple measurement vector (MMV) problem addresses the identification of unknown input vectors that share common sparse support. Even though MMV problems have been traditionally addressed within the context of sensor array signal processing, the recent trend is to apply compressive sensing (CS) due to its capability to estimate sparse support even with an insufficient number of snapshots, in which case classical array signal processing fails. However, CS guarantees the accurate recovery in a probabilistic manner, which often shows inferior performance in the regime where the traditional array signal processing approaches succeed. The apparent dichotomy between the probabilistic CS and deterministic sensor array signal processing has not been fully understood. The main contribution of the present article is a unified approach that revisits the link between CS and array signal processing first unveiled in the mid 1990s by Feng and Bresler. The new algorithm, which we call compressive MUSIC, identifies the parts of support using CS, after which the remaining supports are estimated using a novel generalized MUSIC criterion. Using a large system MMV model, we show that our compressive MUSIC requires a smaller number of sensor elements for accurate support recovery than the existing CS methods and that it can approach the optimal l_0-bound with finite number of snapshots even in cases where the signals are linearly dependent. [ABSTRACT FROM PUBLISHER]