This paper deals with fault isolation in nonlinear analog circuits with tolerance under an insufficient number of independent voltage measurements. A neural network-based L1-norm optimization approach is proposed and utilized in locating the most likely faulty elements in nonlinear circuits. The validity of the proposed method is verified by both extensive computer simulations and practical examples. One simulation example is presented in the paper.
|Title of host publication||Procs IEEE Int Symposium on Circuits & Systems|
|Subtitle of host publication||ISCAS 2001|
|Publisher||Institute of Electrical and Electronics Engineers (IEEE)|
|Publication status||Published - 2001|