Generalized expectation consistent signal recovery for nonlinear measurements
Authors:
- 1 June 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2333-2337
Abstract
In this paper, we propose a generalized expectation consistent signal recovery algorithm to estimate the signal x from the nonlinear measurements of a linear transform output z = Ax. This estimation problem has been encountered in many applications, such as communications with front-end impairments, compressed sensing, and phase retrieval. The proposed algorithm extends the prior art called generalized turbo signal recovery from a partial discrete Fourier transform matrix A to a class of general matrices. Numerical results show the excellent agreement of the proposed algorithm with the theoretical Bayesian-optimal estimator derived using the replica method.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Orthogonal AMPIEEE Access, 2017
- Vector approximate message passing for the generalized linear model2016
- Bayes-Optimal Joint Channel-and-Data Estimation for Massive MIMO With Low-Precision ADCsIEEE Transactions on Signal Processing, 2015
- Turbo Compressed Sensing with Partial DFT Sensing MatrixIEEE Signal Processing Letters, 2014
- Support Recovery With Sparsely Sampled Free Random MatricesIEEE Transactions on Information Theory, 2013
- Probabilistic reconstruction in compressed sensing: algorithms, phase diagrams, and threshold achieving matricesJournal of Statistical Mechanics: Theory and Experiment, 2012
- Generalized approximate message passing for estimation with random linear mixing2011
- Message-passing algorithms for compressed sensingProceedings of the National Academy of Sciences of the United States of America, 2009
- Bayesian Compressive SensingIEEE Transactions on Signal Processing, 2008
- Adaptive and self-averaging Thouless-Anderson-Palmer mean-field theory for probabilistic modelingPhysical Review E, 2001