The idea of the proposed solution (modified-CS) is to solve a convex relaxation of the following problem: find the signal that satisfies the data constraint and is sparsest outside of T, and obtain sufficient conditions for exact reconstruction using modified-CS.
Abstract:
We study the problem of reconstructing a sparse signal from a limited number of its linear projections when a part of its support is known, although the known part may contain some errors. The “known” part of the support, denoted T, may be available from prior knowledge. Alternatively, in a problem of recursively reconstructing time sequences of sparse spatial signals, one may use the support estimate from the previous time instant as the “known” part. The idea of our proposed solution (modified-CS) is to solve a convex relaxation of the following problem: find the signal that satisfies the data constraint and is sparsest outside of T. We obtain sufficient conditions for exact reconstruction using modified-CS. These are much weaker than those needed for compressive sensing (CS) when the sizes of the unknown part of the support and of errors in the known part are small compared to the support size. An important extension called regularized modified-CS (RegModCS) is developed which also uses prior signal estimate knowledge. Simulation comparisons for both sparse and compressible signals are shown.
TL;DR: This paper derives two sparse Bayesian learning algorithms, which have superior recovery performance compared to existing algorithms, especially in the presence of high temporal correlation, and provides analysis of the global and local minima of their cost function.
TL;DR: This paper takes some first steps in the direction of solving inference problems-such as detection, classification, or estimation-and filtering problems using only compressive measurements and without ever reconstructing the signals involved.
TL;DR: In this article, the authors provide essential knowledge and useful tips and tricks that wireless communication researchers need to know when designing CS-based wireless systems, including basic setup, sparse recovery algorithm, and performance guarantee.
TL;DR: In this article, the authors provide a magazine-style overview of the entire field of robust subspace learning (RSL) and tracking (RST) for long data sequences, where the authors assume that the data lies in a low-dimensional subspace that can change over time, albeit gradually.
TL;DR: In this paper, the authors study the recovery conditions of weighted l1 minimization for signal reconstruction from compressed sensing measurements when partial support information is available, and they show that if at least 50% of the (partial) information is accurate, then weighted l 1 minimization is stable and robust under weaker sufficient conditions than the analogous conditions for standard l1 minimizing.
TL;DR: This final installment of the paper considers the case where the signals or the messages or both are continuously variable, in contrast with the discrete nature assumed until now.
TL;DR: In this article, the focus is on recognizing convex optimization problems and then finding the most appropriate technique for solving them, and a comprehensive introduction to the subject is given. But the focus of this book is not on the optimization problem itself, but on the problem of finding the appropriate technique to solve it.
TL;DR: It is possible to design n=O(Nlog(m)) nonadaptive measurements allowing reconstruction with accuracy comparable to that attainable with direct knowledge of the N most important coefficients, and a good approximation to those N important coefficients is extracted from the n measurements by solving a linear program-Basis Pursuit in signal processing.
TL;DR: In this paper, the authors considered the model problem of reconstructing an object from incomplete frequency samples and showed that with probability at least 1-O(N/sup -M/), f can be reconstructed exactly as the solution to the lscr/sub 1/ minimization problem.
Q1. What contributions have the authors mentioned in the paper "Modified-cs: modifying compressive sensing for problems with partially known support" ?
The authors study the problem of reconstructing a sparse signal from a limited number of its linear projections when a part of its support is known, although the known part may contain some errors. The idea of their proposed solution ( modified-CS ) is to solve a convex relaxation of the following problem: find the signal that satisfies the data constraint and is sparsest outside of.
Q2. What is the way to use RegModCS?
RegModCS is useful when exact reconstruction does not occur—either is too small for exact reconstruction or the signal is compressible.
Q3. What is the MAP solution for the i.i.d. training data?
Modified-CS used when and increased it to when (a) , , (b) , , .to on it, then the solution of (32) will be the causal MAP solution under that model.
Q4. What is the smallest real number for a matrix?
The -restricted isometry constant [9], , for a matrix, , is defined as the smallest real number satisfying(1)for all subsets of cardinality and all real vectors of length .
Q5. What is the procedure for generating the random-Gaussian matrix?
2) Generate the random-Gaussian matrix, (generate an matrix with independent identically distributed (i.i.d.) zero mean Gaussian entries and normalize each column to unit norm).
Q6. What is the corresponding DFT for a MRI?
For MRI, is a partial Fourier matrix, i.e., where is an mask which contains a single 1 at a different randomly selected location in each row and all other entries are zero and is the matrix corresponding to the 2-D discrete Fourier transform (DFT).
Q7. What is the significance of comparing the two conditions?
Sufficient conditions for an algorithm serve as a designer’s tool to decide the number of measurements needed for it and in that sense comparing the two sufficient conditions is meaningful.
Q8. How did the authors find a solution to CS?
In practice though, at least with random Gaussian measurements and small enough noise, (6) did turn out to be feasible, i.e., the authors were able find a solution, in all their simulations.
Q9. What is the main question for recursive reconstruction of signal sequences from noisy measurements?
A more important question for recursive reconstruction of signal sequences from noisy measurements, is the stability of the error over time (i.e., how to obtain a time-invariant and small bound on the error over time).