Analysis of the practical capacity of multi-valued hetero-associator considering fault tolerance

Presents a method of pattern recognition using the multi-valued polynomial bidirectional hetero-associator (PBHA). This network can be used for the industrial application of optical character recognition. According to detailed simulations, the PBHA has a higher capacity for pattern pair storage than that of the conventional bidirectional associative memories and fuzzy memories. Meanwhile, the practical capacity of a PBHA considering fault tolerance is discussed. The fault tolerance requirement leads to the discovery of the attraction radius of the basin for each stored pattern pair. The PBHA takes advantage of multi-valued characteristics in evolution equations such that the signal-noise-ratio is significantly increased. We apply the result of this research to pattern recognition problems. The practical capacity of the multi-valued data recognition using the PBHA considering fault tolerance in the worst case is also estimated. Simulation results are presented to verify the derived theory