HDL design of pulse density neural network using simultaneous perturbation

In this paper, we present HDL design of a pulse density neural network circuit with learning ability. We use the simultaneous perturbation method as a learning rule. The learning rule requires only twice forward operations of networks. Thus, without complicated circuit that calculates gradients of an error function, we could construct the network system with learning ability. We designed a pulse density neural network using VHDL. At the same time, we confirmed simulation results of the design through the exclusive OR problem and simple function learning problem.