A HYBRID CELLULAR-AUTOMATON NEURAL-NETWORK CLASSIFIER FOR MULTIVALUEDPATTERNS AND ITS VLSI IMPLEMENTATION

A multi-valued pattern classifier with high discrimination sensitivity and its VLSI implementation proposal on a single chip are presented i n this paper. The classification scheme is based on the combination of a reconfigurable Cellular Automaton and a Neural Network architecture . A 2-D Reconfigurable Hybrid Additive Cellular Automaton (RHACA) arch itecture amplifies the Hamming distance between patterns, whereas a ne ural network architecture, implemented in digital form, assigns vector s of weighing coefficients which take into account the relative signif icance of the sites on the 2-D lattice. The proposed classifier is abl e to operate successfully even for pattern classes of small difference , or for patterns that lie on the decision boundaries between classes. If the training and processing phases are not partitioned, the propos ed classification scheme is able to operate in partially exposed envir onments. With the proper setting of admittance levels into the classes of multi-valued patterns involved, the proposed classifier can also o perate on patterns with partly missing data. The proposed multi-valued pattern classifier can be realized on a single VLSI chip with dimensi ons 7.73 mm x 8.14 mm = 62.96 mm(2) and the expected frequency of oper ation for the chip is 50 MHz.