Prefixless $q$ -Ary Balanced Codes With Fast Syndrome-Based Error Correction

Abstract

We investigate a Knuth-like scheme for balancing $q$ -ary code words, which has the virtue that lookup tables for coding and decoding the prefix are avoided by using precoding and error correction techniques. We show how the scheme can be extended to allow for error correction of single channel errors using a fast decoding algorithm that depends on syndromes only, making it considerably faster compared with the prior art exhaustive decoding strategy. A comparison between the new and prior art schemes, both in terms of redundancy and error performance, completes the study.