A Perceptron Neural Network for Asymmetric Comparison-Based System-Level Fault Diagnosis

The system-level fault diagnosis problem aims at answering the very simple question "Who's faulty and who's fault-free?", in systems known to be diagnosable. In this paper, we answer such a question using neural networks. Our objective is to identify faulty nodes based on an input syndrome that has been generated using the asymmetric comparison model. In such a model, the system, which is composed of interconnected independentheterogeneous nodes, is modeled using an undirected  comparison graph. Tasks are assigned to pairs of nodes and the results of executing these tasks are compared. Based on the agreements and disagreements among the nodes' outputs, the diagnosis algorithm must identify faulty nodes. In general, it is assumed that faults arepermanent, and that at most t nodes can fail simultaneously. The new solution we introduce in this paper uses a perceptron neural network to solve the fault identification problem. The neural network is first trained using various input syndromes with known fault sets. Extensive simulations have been conducted next using randomly generated diagnosable systems. Surprisingly, the neural network was able to identify all the  millions of faulty situations we have tested, including those that are unlikely to occur. Simulations results indicate that the perceptron-based diagnosis algorithm is a viable addition to present diagnosis problems.

[1]  Azzedine Boukerche,et al.  Diagnosing mobile ad-hoc networks: two distributed comparison-based self-diagnosis protocols , 2006, MobiWac '06.

[2]  Amiya Nayak,et al.  System-Level Fault Diagnosis Using Comparison Models: An Artificial-Immune-Systems-Based Approach , 2006, J. Networks.

[3]  Miroslaw Malek,et al.  The consensus problem in fault-tolerant computing , 1993, CSUR.

[4]  Stefano Chessa,et al.  Comparison-based system-level fault diagnosis in ad hoc networks , 2001, Proceedings 20th IEEE Symposium on Reliable Distributed Systems.

[5]  Qing He,et al.  Fault Diagnosis of Induction Motor using Neural Networks , 2007, 2007 International Conference on Machine Learning and Cybernetics.

[6]  Kyung-Yong Chwa,et al.  Schemes for Fault-Tolerant Computing: A Comparison of Modularly Redundant and t-Diagnosable Systems , 1981, Inf. Control..

[7]  M. John,et al.  Antenna Optimization With a Computationally Efficient Multiobjective Evolutionary Algorithm , 2009, IEEE Transactions on Antennas and Propagation.

[8]  Zhi-Hua Zhou,et al.  Using neural networks for fault diagnosis , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[9]  Kang G. Shin,et al.  Probabilistic diagnosis of multiprocessor systems , 1994, CSUR.

[10]  Azzedine Boukerche,et al.  A distributed fault identification protocol for wireless and mobile ad hoc networks , 2008, J. Parallel Distributed Comput..

[11]  Miroslaw Malek,et al.  A comparison connection assignment for diagnosis of multiprocessor systems , 1980, ISCA '80.

[12]  Abhijit Sengupta,et al.  On self-diagnosable multiprocessor systems: diagnosis by the comparison approach , 1989, [1989] The Nineteenth International Symposium on Fault-Tolerant Computing. Digest of Papers.

[13]  W. Cholewa,et al.  Fault Diagnosis: Models, Artificial Intelligence, Applications , 2004 .

[14]  Mourad ELHADEF,et al.  Efficient comparison-based fault diagnosis of multiprocessor systems using an evolutionary approach , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[15]  Yuan Yan Tang,et al.  Efficient Fault Identification of Diagnosable Systems under the Comparison Model , 2007, IEEE Transactions on Computers.

[16]  Andrzej Pelc,et al.  Complexity of Fault Diagnosis in Comparison Models , 1992, IEEE Trans. Computers.

[17]  Douglas M. Blough,et al.  The Broadcast Comparison Model for On-Line Fault Diagnosis in Multicomputer Systems , 1999, IEEE Trans. Computers.

[18]  Andrzej Pelc,et al.  Undirected Graph Models for System-Level Fault Diagnosis , 1991, IEEE Trans. Computers.

[19]  GERNOT METZE,et al.  On the Connection Assignment Problem of Diagnosable Systems , 1967, IEEE Trans. Electron. Comput..