Diagnosis of specification parametric faults in the FPAA: the RBF neural network approach
Damian E. Grzechca, Tomasz Golonek, Jerzy Rutkowski · 2008
This paper presents an idea of the parametric fault detection and location in the field programmable analog arrays (FPAA) with use of the time domain input excitation, and Radial Basis Function (RBF) Neural Network (NN) application. The aim of the work is to diagnose a parametric fault in the CAB based on the output time domain response. A group of selected features feed the input layer of the neural network and the output indicates state of the examined FPAA block. The paper gives brief information about fault diagnosis in the analog electronic circuit as well as practical implementation of proposed method. Exemplary circuit is investigated and results with conclusions are discussed.