Submitted to Ieee Transactions on Information Theory 2 I

Minimization of the `1 (or maximum) norm subject to a constraint that imposes consistency to an underdetermined system of linear equations finds use in a large number of practical applications, including vector quantization, peak-to-average power ratio (PAPR) (or “crest factor”) reduction in communication systems, approximate nearest neighbor search, and peak force minimization in robotics and control. This paper analyzes the fundamental properties of signal representations obtained by solving such a convex optimization problem. We develop bounds on the maximum magnitude of such representations using the uncertainty principle (UP) introduced by Lyubarskii and Vershynin, IEEE Trans. IT, 2010, and study the efficacy of `1-norm-based PAPR reduction. Our analysis shows that matrices satisfying the UP, such as randomly subsampled Fourier or i.i.d. Gaussian matrices, enable the computation of what we call democratic representations, whose entries all have small and similar magnitude, as well as low PAPR. To compute democratic representations at low computational complexity, we present two new, efficient convex optimization algorithms. We finally demonstrate the efficacy of democratic representations for PAPR reduction in a DVB-T2-based broadcast system.