A single-layer asymmetric RNN with low hardware complexity for solving linear equations

Abstract

A single layer neural network for the solution of linear equations is presented. The proposed circuit is based on the standard Hopfield model albeit with the added flexibility that the interconnection weight matrix need not be symmetric. This results in an asymmetric Hopfield neural network capable of solving linear equations. PSPICE simulation results are given which verify the theoretical predictions. A simple technique to incorporate re-configurability into the circuit for setting the different weights of the interconnection is also included. Experimental results for circuits set up to solve small problems further confirm the operation of the proposed circuit.