Recursive sparse reconstruction
AssigneeIowa State University Research Foundation, Inc.
InventorNamrata Vaswani
Life of this patent
Estimating the term…
Abstract
A method for real-time reconstruction is provided. The method includes receiving a sparse signal sequence one at a time and performing compressed sensing on the sparse signal sequence in a manner which causally estimates a time sequence of spatially sparse signals and generates a real-time reconstructed signal. Recursive algorithms provide for causally reconstructing a time sequence of sparse signals from a greatly reduced number of linear projection measurements.
Claims30
A method for real-time reconstruction, the method comprising:
- receiving a sparse signal sequence one at a time;
- and
- performing compressed sensing on the sparse signal sequence in a manner which casually estimates a time sequence of spatially sparse signals and generates a real-time reconstructed signal.
An apparatus for providing real-time reconstruction of a time-sequence of sparse signals, the apparatus comprising:
- at least one sensor for sensing a sparse signal sequence;
- a processor configured to perform compressed sensing on the sparse signal sequence in a manner which casually estimates a time sequence of spatially sparse signals and generates a real-time reconstructed signal.
A method for performing Kalman filtered compressed sensing, comprising:
- receiving a sparse signal sequence one at a time;
- estimating a support set using compressed sensing for a first sparse signal within the sparse signal sequence;
- applying a reduced order Kalman filter with the support set to subsequent sparse signals within the sparse signal sequence;
- estimating additions to the support set using compressed sensing and updating the support set.
A method for performing least squares compressed sensing, comprising:
- receiving a sparse signal sequence one at a time;
- estimating a support set using compressed sensing for a first sparse signal within the sparse signal sequence;
- applying least squares estimation with the support set to subsequent sparse signals within the sparse signal sequence;
- estimating additions to the support set using compressed sensing and updating the support set.
A method for performing modified compressed sensing, comprising:
- receiving a sparse signal;
- determining a partially known support set, T;
- performing a reconstruction of a signal from the sparse signal by solving a problem both satisfying a data constraint and having a support set containing a smallest number of new additions to the partially known support set, T.
Dependencies are inferred from the claim text.