from Phys. Rev. X 2, 021005 (2012)

by Brian Busemeyer

Compressed sensing

Compressed sensing is a way of extracting a full signal out of a sparse sampling. It's only requirement is that the signal has a sparse representation in some basis, which is actually true for most interesting signals that we encounter.

Presentation Summary

In this presentation, I present:

  • the basic problem this solves.
  • why it makes sense to optimize for sparsity.
  • results from my own implementation of the l-1 minimization approach.
  • an exploration of the parameter space for which this method is successful.
  • recent developments in the field, and it’s connection to physics.

Examples

My compressed sensing notebook (html) and related python library.

References

Original paper (I think? in some sense?):

IEEE Trans. Inf. Theory 52, 1289 (2006)

Probabilistic seeding:

Phys. Rev. X 2, 021005 (2012)

Simultaneous measurement of physical observables.:

Phys. Rev. Lett. 112, 253602 (2014)

All signal processing.

ALGORITHM
signal processing numerical method

Archive