A combined clustering and neural network approach for analog multiple hardfault classification

A new neural network-based fault classification strategy for hard multiple faults in analog circuits is proposed. The magnitude of the harmonics of th e Fourier components of the circuit response at different test nodes due to a sinusoidal input signal are first measured or simulated. A selection cri terion for determining the best components that describe the circuit behavi our under fault-free (nominal) and fault situations is presented. An algori thm that estimates the overlap between different faults in the measurement space is also introduced. The learning vector quantization neural network i s then effectively trained to classify circuit faults. Performance measures reveal very high classification accuracy in both training and testing stag es. Two different examples, which demonstrate the proposed strategy, are de scribed.