Asymptotic MMSE analysis under sparse representation modeling

Research output: Contribution to journalArticlepeer-review


Compressed sensing is a signal processing technique in which data is acquired directly in a compressed form. There are two modeling approaches that can be considered: the worst-case (Hamming) approach and a statistical mechanism, in which the signals are modeled as random processes rather than as individual sequences. In this paper, the second approach is studied. In particular, we consider a model of the form Y=HX+W, where each comportment of X is given by Xi=SiUi, where {Ui} are independent and identically distributed (i.i.d.) Gaussian random variables, {Si} are binary i.i.d. random variables independent of {Ui}, H∈Rk×n is a random matrix with i.i.d. entries, and W is white Gaussian noise. Using a direct relationship between optimum estimation and certain partition functions, and by invoking methods from statistical mechanics and from random matrix theory (RMT), we derive an asymptotic formula for the minimum mean-square error (MMSE) of estimating the input vector X given Y and H, as k,n→∞, keeping the measurement rate, R=k/n, fixed. In contrast to previous derivations, which are based on the replica method, the analysis carried out in this paper is rigorous.

Original languageEnglish
Pages (from-to)320-332
Number of pages13
JournalSignal Processing
StatePublished - 1 Feb 2017


  • Compressed sensing (CS)
  • Conditional mean estimation
  • Minimum mean-square error (MMSE)
  • Partition function
  • Phase transitions
  • Random matrix
  • Replica method
  • Statistical-mechanics
  • Threshold effect

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering


Dive into the research topics of 'Asymptotic MMSE analysis under sparse representation modeling'. Together they form a unique fingerprint.

Cite this