Dynamic Iterative Pursuit

@article{Zachariah2012DynamicIP,
  title={Dynamic Iterative Pursuit},
  author={Dave Zachariah and Saikat Chatterjee and M. Jansson},
  journal={IEEE Transactions on Signal Processing},
  year={2012},
  volume={60},
  pages={4967-4972}
}
For compressive sensing of dynamic sparse signals, we develop an iterative pursuit algorithm. A dynamic sparse signal process is characterized by varying sparsity patterns over time/space. For such signals, the developed algorithm is able to incorporate sequential predictions, thereby providing better compressive sensing recovery performance, but not at the cost of high complexity. Through experimental evaluations, we observe that the new algorithm exhibits a graceful degradation at… 

Figures from this paper

Distributed predictive subspace pursuit
TLDR
In a compressed sensing setup with jointly sparse, correlated data, a distributed greedy algorithm called distributed predictive subspace pursuit is developed, based on estimates from neighboring sensor nodes, which provides better performance than current state-of-the-art algorithms.
Variational Bayesian dynamic compressive sensing
TLDR
This paper proposes a computationally efficient mean-field variational Bayes algorithm to learn the dynamic compressed sensing model without parameter tuning, and considers both the online and offline scenarios.
Online convex optimization meets sparsity
TLDR
A new perspective on tracking time-varying sparse signals is proposed, based on the theory on online convex optimization, which has been developed in the machine learning community, and a strongly convex model is exploited.
Methods for Distributed Compressed Sensing
TLDR
An overview of the current literature regarding distributed compressed sensing is provided, in particular, aspects of network topologies, signal models and recovery algorithms are discussed.
Dynamic Filtering of Time-Varying Sparse Signals via $\ell _1$ Minimization
TLDR
Two algorithms for dynamic filtering of sparse signals that are based on efficient ℓ1 optimization methods that represent the first strong performance analysis of dynamic filtering algorithms for time-varying sparse signals as well as state-of-the-art performance in this emerging application.
Re-Weighted l_1 Dynamic Filtering for Time-Varying Sparse Signal Estimation
TLDR
This work presents a re-weighted l_1 dynamic filtering scheme for causal signal estimation that utilizes both sparsity assumptions and dynamic structure and incorporates both dynamic and sparsity priors in the estimation procedure in a robust and efficient algorithm.
Distributed greedy pursuit algorithms
Design and Analysis of a Greedy Pursuit for Distributed Compressed Sensing
TLDR
This paper develops a distributed parallel pursuit (dipp) algorithm based on exchange of information about estimated support-sets at sensors, which converges to a performance level that depends on a scaled additive measurement noise power (convergence in theory) where the scaling coefficient is a function of RIP parameters and information processing quality parameters.
Greedy Algorithms for Distributed Compressed Sensing
Compressed sensing (CS) is a recently invented sub-sampling technique that utilizes sparsity in full signals. Most natural signals possess this sparsity property. From a sub-sampled vector, some CS
Sparse Bayesian Learning With Dynamic Filtering for Inference of Time-Varying Sparse Signals
TLDR
Numerical simulations show that SBL-DF converges much faster and to more accurate solutions than standard SBL and other dynamical filtering algorithms, and outperforms state of the art algorithms when the dictionary contains the challenging coherence and column scaling structure found in many practical applications.
...
1
2
3
...

References

SHOWING 1-10 OF 23 REFERENCES
Robust matching pursuit for recovery of Gaussian sparse signal
TLDR
A minimum mean square error (MMSE) estimation based iterative greedy search algorithm is developed using a priori statistical knowledge that provides a robust CS reconstruction performance compared to an existing least square based algorithm.
Subspace Pursuit for Compressive Sensing Signal Reconstruction
TLDR
The presented analysis shows that in the noiseless setting, the proposed algorithm can exactly reconstruct arbitrary sparse signals provided that the sensing matrix satisfies the restricted isometry property with a constant parameter.
Look ahead orthogonal matching pursuit
TLDR
The effect of look ahead strategy is shown to provide a significant improvement in performance in the recovery performance of the existing orthogonal matching pursuit (OMP) algorithm.
Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit
This paper demonstrates theoretically and empirically that a greedy algorithm called orthogonal matching pursuit (OMP) can reliably recover a signal with m nonzero entries in dimension d given O(m ln
Uniform Uncertainty Principle and Signal Recovery via Regularized Orthogonal Matching Pursuit
TLDR
This paper finds a simple regularized version of Orthogonal Matching Pursuit (ROMP) which has advantages of both approaches: the speed and transparency of OMP and the strong uniform guarantees of L1-minimization.
Kalman filtered Compressed Sensing
  • N. Vaswani
  • Computer Science
    2008 15th IEEE International Conference on Image Processing
  • 2008
TLDR
This work considers the problem of reconstructing time sequences of spatially sparse signals from a limited number of linear "incoherent" measurements, in real-time, and uses Compressed Sensing to estimate the support set of the initial signal's transform vector.
Sparse signal recovery in the presence of correlated multiple measurement vectors
  • Zhilin Zhang, B. Rao
  • Computer Science
    2010 IEEE International Conference on Acoustics, Speech and Signal Processing
  • 2010
TLDR
This work model sources as AR processes, and then incorporate such information into the framework of sparse Bayesian learning for sparse signal recovery, and proposes algorithms to account for temporal correlation and its impact on performance.
CoSaMP: Iterative signal recovery from incomplete and inaccurate samples
LS-CS-Residual (LS-CS): Compressive Sensing on Least Squares Residual
  • N. Vaswani
  • Computer Science
    IEEE Transactions on Signal Processing
  • 2010
TLDR
This work bound CS-residual error and shows “stability” of LS-CS over time for a signal model that allows support additions and removals, and that allows coefficients to gradually increase (decrease) until they reach a constant value (become zero).
Methods for Sparse Signal Recovery Using Kalman Filtering With Embedded Pseudo-Measurement Norms and Quasi-Norms
TLDR
Two algorithms that rely on a Kalman filter endowed with a pseudo-measurement (PM) equation to solve the theory of compressed sensing's convex constrained minimization problem are proposed.
...
1
2
3
...