On Self-Dual Quantum Codes, Graphs, and Boolean Functions

TL;DR

This paper explores the representation of self-dual quantum codes through graphs and Boolean functions, classifies codes up to length 12, and investigates cryptographic properties and entanglement measures related to these codes.

Contribution

It introduces a classification of self-dual additive quantum codes using graph operations and Boolean functions, extending known results to length 12 and beyond, and links cryptographic properties to quantum code parameters.

Findings

Classified all inequivalent self-dual additive codes over GF(4) up to length 12.

Established a connection between quantum code distance and Boolean function propagation criteria.

Proposed methods to construct non-quadratic Boolean functions with low PAR_IHN.

Abstract

A short introduction to quantum error correction is given, and it is shown that zero-dimensional quantum codes can be represented as self-dual additive codes over GF(4) and also as graphs. We show that graphs representing several such codes with high minimum distance can be described as nested regular graphs having minimum regular vertex degree and containing long cycles. Two graphs correspond to equivalent quantum codes if they are related by a sequence of local complementations. We use this operation to generate orbits of graphs, and thus classify all inequivalent self-dual additive codes over GF(4) of length up to 12, where previously only all codes of length up to 9 were known. We show that these codes can be interpreted as quadratic Boolean functions, and we define non-quadratic quantum codes, corresponding to Boolean functions of higher degree. We look at various cryptographic properties of Boolean functions, in particular the propagation criteria. The new aperiodic propagation criterion (APC) and the APC distance are then defined. We show that the distance of a zero-dimensional quantum code is equal to the APC distance of the corresponding Boolean function. Orbits of Boolean functions with respect to the {I,H,N}^n transform set are generated. We also study the peak-to-average power ratio with respect to the {I,H,N}^n transform set (PAR_IHN), and prove that PAR_IHN of a quadratic Boolean function is related to the size of the maximum independent set over the corresponding orbit of graphs. A construction technique for non-quadratic Boolean functions with low PAR_IHN is proposed. It is finally shown that both PAR_IHN and APC distance can be interpreted as partial entanglement measures.