Neural network for the exact RLS adaptive algorithm (Q1316116)

A recursive least squares (RLS) algorithm based on a modification of the Hopfield neural network model is proposed to implement exact RLS algorithm in real time adaptive signal processing. The modification involves the selection of a suitale relationship function from neurons. The stability of the so constructed neural network is shown. Results are provided arbitrarily close to the accurate optimal solution of the RLS algorithm within an elapsed time of only a few time constants of the network. The interconnection strengths and bias currents can be obtained from the available data with a few computations (much fewer than in the exact RLS algorithm).