Artificial Neural Networks have proven to be an effective tool in performing
complex functions in a variety fields. These include pattern recognition,
classification, vision, control systems and prediction. In this research we present a
method for implementing an analog PID controller as an application of using
FPAA technology. Also we have used Field Programmable Analog Arrays
(FPAAs) technology to realize hardware implementation of an artificial neural
network. We describe two models of artificial neuron implemented in hardware
by using FPAA. A simplified realization for neurons with hyperbolic tangent
activation function is used to reduce the complexity of the neural network
architecture.
The built neural network was tested with the two-input XOR problem. Also we
have used twelve FPAA chips to realize a multilayer neural network to solve
"TCLX" character recognition problem.