TY - GEN
T1 - On multiple-fault diagnosis of baseline interconnection networks
AU - Lin, Shun Shii
AU - Lin, Ferng Ching
N1 - Publisher Copyright:
© 1992 IEEE.
PY - 1992
Y1 - 1992
N2 - We use the general fault model of [7] to study the multiple-fault diagnosis method for the baseline interconnection networks. We find that six tests are sufficient for the 4∗4 and 8∗8 network diagnosis plus that the number of tests is double as the network size increases. A new and systematic multiple-fault diagnosis procedure is developed. The following new results are derived in this study: 1) The necessary and sufficient conditions of test vectors for detecting a fault or multiple faults and determining the fault types in a switching element or a link and in a 4∗4 baseline network are derived. 2) Six tests are necessary and sufficient for detecting multiple faults and determine partially the fault types in a 4∗4 baseline network and an 8∗8 baseline network. Multiple faults can be detected and the fault types can be partially determined by 2log2N tests in an N∗N baseline network, N ≥ 8. 4) If K distinct vectors are sufficient for diagnosing a R∗R baseline network, then d[log2(N∗K/R+2)] tests are sufficient for diagnosing an N∗N baseline network for N ≥ R.
AB - We use the general fault model of [7] to study the multiple-fault diagnosis method for the baseline interconnection networks. We find that six tests are sufficient for the 4∗4 and 8∗8 network diagnosis plus that the number of tests is double as the network size increases. A new and systematic multiple-fault diagnosis procedure is developed. The following new results are derived in this study: 1) The necessary and sufficient conditions of test vectors for detecting a fault or multiple faults and determining the fault types in a switching element or a link and in a 4∗4 baseline network are derived. 2) Six tests are necessary and sufficient for detecting multiple faults and determine partially the fault types in a 4∗4 baseline network and an 8∗8 baseline network. Multiple faults can be detected and the fault types can be partially determined by 2log2N tests in an N∗N baseline network, N ≥ 8. 4) If K distinct vectors are sufficient for diagnosing a R∗R baseline network, then d[log2(N∗K/R+2)] tests are sufficient for diagnosing an N∗N baseline network for N ≥ R.
UR - http://www.scopus.com/inward/record.url?scp=85023946438&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85023946438&partnerID=8YFLogxK
U2 - 10.1109/PCCC.1992.200590
DO - 10.1109/PCCC.1992.200590
M3 - Conference contribution
AN - SCOPUS:85023946438
T3 - 11th Annual International Phoenix Conference on Computers and Communication, IPCCC 1992 - Proceedings
SP - 448
EP - 455
BT - 11th Annual International Phoenix Conference on Computers and Communication, IPCCC 1992 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th Annual International Phoenix Conference on Computers and Communication, IPCCC 1992
Y2 - 1 April 1992 through 3 April 1992
ER -