Accession Number : ADA291219

Title :   Robust Testing of Cellular Neural Networks.

Descriptive Note : Technical rept.,

Corporate Author : TEXAS A AND M UNIV COLLEGE STATION DEPT OF ELECTRICAL ENGINEERING

Personal Author(s) : Pineda de Gyvez, Jose

PDF Url : ADA291219

Report Date : 01 FEB 1995

Pagination or Media Count : 19

Abstract : A method for detecting circuit faults within two dimensional Cellular Neural Network (CNN) arrays is presented. The need to develop robust methods for detecting faults is driven by the lack of visibility of the internal nodes in VLSI implementations of CNN arrays. The method is composed of both behavioral and parametric tests and detects faults independent of the size or topology of the CNN array. The behavioral tests reveal nodes that exhibit opened, shorted, or stuck-at a supply voltage faults. The parametric tests reveal excessive time constant mismatches, impedance mismatches and voltage offsets. Seven fault cases are introduced into a macromodel of a voltage mode CNN array to provide insight to the usefulness of the proposed testing methodology.

Descriptors :   *IMAGE PROCESSING, *NEURAL NETS, *SUPERCOMPUTERS, COMPUTERIZED SIMULATION, PARAMETRIC ANALYSIS, COMPUTER AIDED DESIGN, REAL TIME, TWO DIMENSIONAL, CELLS, TEST METHODS, ARRAYS, VERY LARGE SCALE INTEGRATION, PROTOTYPES, VOLTAGE, NODES, TOPOLOGY, INTERNAL, VISIBILITY, BEHAVIOR, CIRCUITS, PSYCHOLOGICAL TESTS, SUPPLIES, FAULTS.

Subject Categories : Computer Hardware
      Computer Systems
      Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE