Integer Codes Correcting Single Errors within Two Bytes

Abstract

This paper presents a class of integer codes suitable for use in optical networks with low error rates. The presented codes are constructed with the help of a computer and have two important features: first, they can correct single errors affecting one or two b-bit bytes, and second, they use processor-friendly operations to encode/decode data bits. The effectiveness of the presented codes is demonstrated on theoretical models of four-core and six-core processors. The obtained results show that the decoder throughput reaches 14.70Gbps, which is above the operating speed of 10G networks. Finally, the paper compares the proposed codes with BCH codes of similar properties. The comparison is made in terms of redundancy and the number of decoding operations.