The CBR-and-MAS technique is introduced into Diagnosis Systems for Distributed Faults and a Diagnosis Systems Model for CBR-and-MAS Based Distributed Faults is proposed.

A method of the system-level fault diagnosis based on the non-idempotent linear equations;

In the context of system-level fault diagnosis of multiprocessor systems,one major issue is to identify a set of processors that are most probable to fail.

In this paper a method based on inhomogeneous linear equation system is presented for test model of system-level fault diagnosis,the paper gives equations of the PMC model,BGM model,Chwa & Hakimi model,Maleks model,and further gives optimal diagnosis s algorithm and its C++ algorithm description.

This article defines the strong t diagnosable system of PMC model based on mutual test, gives a series of characters and theorems and puts forward the theory of grouping algorithm for the system level fault diagnosis,so the complexity of system level fault diagnosis is greatly decrease

The concept of body was firstly applied to Chaw & Hakimi model in system level fault diagnosis.

This paper presents and discusses some generating principles and methods of the test graph on the system level fault diagnosis about random test graph,limit t fault diagnosable test graph, t fault diagnosable test graph and so on.

Elhadef and Ayeb devised a genetic algorithm for the system-level diagnosis of multi- computers,where the fitness function is calculated by comparing the given syndrome with the syndrome randomly produced by the current guess fault set.

The t/k-diagnosis strategy of system-level diagnosis significantly increases the self-diagnosing capability of multiprocessor systems at the cost of at most k fault-free nodes being incorrectly identified as faulty.