A new adaptive architecture: analogue synthesiser of orthogonal functions

摘要: A new adaptive nonlinear (neural-like) architecture, an analogue synthesiser of orthogonal functions which is able to produce a plurality mutually signals as time such Legendre, Chebyshev and Hermite polynomials, cosine basis functions, smoothed basis, etc., proposed. proof-of-concept breadboard version the described. The device characterised by very fast (approximately 100 iterations) stable process signal synthesis. proposed could find applications e.g. in systems function approximation, particular main unit implementation so-called polynomial-based (CPB) neural networks, alternative Volterra polynomial also preprocessing element (performing some transforms, filtration, etc.) network-based information processing.