The maximum-likelihood decoding threshold for graphic codes

TL;DR

This paper establishes an upper bound on the maximum-likelihood decoding threshold for cycle codes of graphs, revealing that regular graphs asymptotically achieve this bound, thus advancing understanding of decoding limits in graph-based codes.

Contribution

It derives a precise upper bound on the decoding threshold for cycle codes and characterizes when this bound is asymptotically tight, specifically for regular graphs.

Findings

Upper bound on decoding threshold: _{ ext{max}}(R)  rac{(1-\u221a{R})^2}{2(1+R)}

Equality in the bound is achieved asymptotically by cycle codes of regular graphs.

Provides insight into the decoding performance limits of graph-based codes.

Abstract

For a class $\mathcal{C}$ of binary linear codes, we write $\theta_{\mathcal{C}}\colon (0,1) \to [0,\frac{1}{2}]$ for the maximum-likelihood decoding threshold function of $\mathcal{C}$, the function whose value at $R \in (0,1)$ is the largest bit-error rate $p$ that codes in $\mathcal{C}$ can tolerate with a negligible probability of maximum-likelihood decoding error across a binary symmetric channel. We show that, if $\mathcal{C}$ is the class of cycle codes of graphs, then $\theta_{\mathcal{C}}(R) \le \frac{(1-\sqrt{R})^2}{2(1+R)}$ for each $R$, and show that equality holds only when $R$ is asymptotically achieved by cycle codes of regular graphs.