JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]


Journal of Information Science and Engineering, Vol. 40 No. 2, pp. 231-244


A New Tree Structure for Local Diagnosis


MEIRUN CHEN1, XIAO-YAN LI2, CHENG-KUAN LIN3,4,5,+
AND KUNG-JUI PAI6
1School of Mathematics and Statistics
Xiamen University of Technology
Xiamen, 361024 P.R. China

2College of Computer and Data Science
Fuzhou University
Fuzhou, 350108 P.R. China

3Department of Computer Science
4Undergraduate Degree Program of Systems Engineering and Technology
National Yang Ming Chiao Tung University
Hsinchu, 30010 Taiwan
E-mail: cklin@nycu.edu.tw
+
5Computer Science and Information Engineering
Chung Cheng Institute of Technology
National Defense University
Taoyuan, 30010 Taiwan

6Department of Industrial Engineering and Management
Ming Chi University of Technology
New Taipei City, 243303 Taiwan


Diagnosability is an important parameter to measure the fault tolerance of a multiprocessor system. If we only care about the state of a node, instead of doing the global diagnosis, Hsu and Tan proposed the idea of local diagnosis. Chiang and Tan provided an extended star structure to diagnose a node under comparison model. In this work, we evaluate the local diagnosability better by proposing a tree structure around this node. We provide the corresponding algorithm to diagnose the node. Simulation results are presented for different failure probability of a node in the tree and different percentage of faulty nodes in the tree, showing the performance of our algorithm.


Keywords: fault diagnosis, comparison model, diagnosis algorithm, local diagnosability, tree

  Retrieve PDF document (JISE_202402_02.pdf)