A Double Error-Correction Scheme for Peripheral Systems

In multiple digital recording peripheral systems, the errors which are encountered are burst-type errors which may occur in several tracks or channels. This paper discusses methods and implementation procedures for correcting a series of errors or erasures of arbitrary length in any one or two tracks of a multi-channel system. We have shown that such errors in multiple tracks may be detected and corrected using a two-redundant Hamming code over a Galois field mathematical structure in conjunction with error track pointers. The detection/correction scheme considered here can correct N-bit blocks, thereby correcting an error burst of arbitrary length in any single or pair of tracks. The error-correction (EC) scheme is shown to be applicable to two types of data formats or arrays. Additional error detection is provided to check the validity of error correction to enhance reliability above that of double error correction. Implementation methods pertaining to the above EC scheme are discussed. Examples of single and double error corrections for two types of data formats or arrays are included.