TY - GEN

T1 - A new construction for n-track (d, k) codes with redundancy

AU - Ke, Ligang

AU - Marcellin, Michael W.

PY - 1994/12/1

Y1 - 1994/12/1

N2 - Digital magnetic and optical storage systems employing NRZI recording use (d, k) codes. The d-parameter specifies the minimum number of 0's occurring between 1's while the k-parameter specifies the maximum number of 0's between l's. The n-track (d,k) codes (denoted as (d,k;n) codes) are extensions of (d, k) codes for use in multiple-track systems. Instead of imposing each track to individually satisfy both constraints, (d,k;n) codes satisfy the d-constraint in each track individually while relaxing the k-constraint by allowing it to be satisfied jointly by the multiple tracks. Although (d,k;n) codes can provide significant capacity increases over (d, k) codes, they suffer from the fact that a single faulty track can cause loss of synchronization and hence, loss of the data on all tracks. Orcutt and Marcellin (see IEEE Trans. on Inform. Theory, Sept., 1993) introduced n-track (d,k) codes with a redundancy of r (denoted as (d,k;n,r) codes) which allow for r faulty tracks by mandating that all subsets of n-r tracks satisfy the joint k-constraint. We propose a new method to construct (d,k; n, r) codes. These codes have simple encoding and decoding schemes, gain a large part of the capacity increase possible when using (d,k; n,r) codes, and are considerably more robust to faulty tracks.

AB - Digital magnetic and optical storage systems employing NRZI recording use (d, k) codes. The d-parameter specifies the minimum number of 0's occurring between 1's while the k-parameter specifies the maximum number of 0's between l's. The n-track (d,k) codes (denoted as (d,k;n) codes) are extensions of (d, k) codes for use in multiple-track systems. Instead of imposing each track to individually satisfy both constraints, (d,k;n) codes satisfy the d-constraint in each track individually while relaxing the k-constraint by allowing it to be satisfied jointly by the multiple tracks. Although (d,k;n) codes can provide significant capacity increases over (d, k) codes, they suffer from the fact that a single faulty track can cause loss of synchronization and hence, loss of the data on all tracks. Orcutt and Marcellin (see IEEE Trans. on Inform. Theory, Sept., 1993) introduced n-track (d,k) codes with a redundancy of r (denoted as (d,k;n,r) codes) which allow for r faulty tracks by mandating that all subsets of n-r tracks satisfy the joint k-constraint. We propose a new method to construct (d,k; n, r) codes. These codes have simple encoding and decoding schemes, gain a large part of the capacity increase possible when using (d,k; n,r) codes, and are considerably more robust to faulty tracks.

UR - http://www.scopus.com/inward/record.url?scp=84894290175&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894290175&partnerID=8YFLogxK

U2 - 10.1109/ISIT.1994.394830

DO - 10.1109/ISIT.1994.394830

M3 - Conference contribution

AN - SCOPUS:84894290175

SN - 0780320158

SN - 9780780320154

T3 - IEEE International Symposium on Information Theory - Proceedings

BT - Proceedings - 1994 IEEE International Symposium on Information Theory, ISIT 1994

T2 - 1994 IEEE International Symposium on Information Theory, ISIT 1994

Y2 - 27 June 1994 through 1 July 1994

ER -