English

Derandomizing compressed sensing with combinatorial design

Information Theory 2018-12-20 v1 Combinatorics math.IT

Abstract

Compressed sensing is the art of reconstructing structured nn-dimensional vectors from substantially fewer measurements than naively anticipated. A plethora of analytic reconstruction guarantees support this credo. The strongest among them are based on deep results from large-dimensional probability theory that require a considerable amount of randomness in the measurement design. Here, we demonstrate that derandomization techniques allow for considerably reducing the amount of randomness that is required for such proof strategies. More, precisely we establish uniform s-sparse reconstruction guarantees for Cslog(n)C s \log (n) measurements that are chosen independently from strength-four orthogonal arrays and maximal sets of mutually unbiased bases, respectively. These are highly structured families of C~n2\tilde{C} n^2 vectors that imitate signed Bernoulli and standard Gaussian vectors in a (partially) derandomized fashion.

Keywords

Cite

@article{arxiv.1812.08130,
  title  = {Derandomizing compressed sensing with combinatorial design},
  author = {Peter Jung and Richard Kueng and Dustin G. Mixon},
  journal= {arXiv preprint arXiv:1812.08130},
  year   = {2018}
}

Comments

12 pages, 2 figures