Signal error-correction system and method
US6154870A · kind A · utility
Assignee
Inventors
Key dates
| Filing date | Jun 4, 1998 |
| Grant date | Nov 28, 2000 |
| Priority date | — |
| Expiry date | Jun 4, 2018 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH03M13/6505
- WIPO fieldBasic communication processes
- WIPO sectorElectrical engineering
Abstract
A Viterbi PRML system and method providing a new code with distance properties such that some tribits are allowed but no longer sequences are allowed. A code rate 8/ 9 is constructed for EPR4 and E.sup.2 PR4 channels and the system independently maps 8-bit blocks of user data to 9-bit channel sequences. The precoder has transfer function, f(D)=1/(1.sym.D), and produces a binary channel input x(D), which is fed to a coder, to provide an output signal y(D), which is transmitted and corrupted by noise. The corrupted signal is received and fed to a Viterbi detector. The signal is decoded to produce an estimate of the 8-bit data bytes, as reconstructed to be freed from noise corruption. The encoding protocol of the invention is implemented in the encoder. The protocol is: no sequence of 4 consecutive transitions occurs in any 9-bit codeword; no 9-bit codeword ends with a sequence of 2 or more consecutive transitions; no 9-bit codeword begins with more than 2 consecutive transitions; and sequences of 3 consecutive transitions, if any, begin only on a 2nd, 4th, 6th, or 9th bit of said 9-bit codeword. The protocol is preferably augmented by an additional constraint: no 9-bit codeword has t…
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.