Synchronization Error-Correction Coding Based on Probabilistic Inference and Sparse Matrices

Abstract

In communication and recording systems affected by uncertainties in timing or time noise, synchronization errors, such as symbol insertions and deletions, can significantly impair the reliability and efficiency of the system. In this article, we review synchronization error-correcting codes based on probabilistic inference and sparse matrices. We first present a probabilistic channel model of synchronization errors, formalize a decoding problem for the channel as probabilistic inference, and show a belief-propagation decoding algorithm. Then, we review some existing synchronization error-correcting codes.