Sparse superposition codes under VAMP decoding with generic rotational invariant coding matrices

TL;DR

This paper extends sparse superposition codes to rotational invariant matrices, analyzes VAMP decoding, and provides design principles for capacity-achieving codes with optimal thresholds.

Contribution

It introduces a generalized coding scheme with rotational invariant matrices, analyzes VAMP decoding performance, and derives a spectral criterion for optimal code design.

Findings

Structured matrices outperform i.i.d. matrices.

VAMP decoding threshold is spectrum-independent.

Spectral criterion guides capacity-achieving code design.

Abstract

Sparse superposition codes were originally proposed as a capacity-achieving communication scheme over the gaussian channel, whose coding matrices were made of i.i.d. gaussian entries.We extend this coding scheme to more generic ensembles of rotational invariant coding matrices with arbitrary spectrum, which include the gaussian ensemble as a special case. We further introduce and analyse a decoder based on vector approximate message-passing (VAMP).Our main findings, based on both a standard replica symmetric potential theory and state evolution analysis, are the superiority of certain structured ensembles of coding matrices (such as partial row-orthogonal) when compared to i.i.d. matrices, as well as a spectrum-independent upper bound on VAMP's threshold. Most importantly, we derive a simple "spectral criterion " for the scheme to be at the same time capacity-achieving while having the best possible algorithmic threshold, in the "large section size" asymptotic limit. Our results therefore provide practical design principles for the coding matrices in this promising communication scheme.