COMPLEX SIGNAL PREDICTION BASED ON DYNAMIC SYSTEM ALLOCATION LIMITS

I. . Sidorkina, A. . Egoshin, D. . Shumkov


Read the full article 

Abstract

 

The paper proposes two methods for predicting complex signal, based on multilayer perceptron and SVM,
implemented in the software complex. The training set for the predictor is based on the allocation of border
implementations of dynamic systems in the signal.

Keywords: forecasting, change-point detection, fractal, dynamic systems, neural networks, SVM
Copyright 2001-2017 ©
Scientific and Technical Journal
of Information Technologies, Mechanics and Optics.
All rights reserved.

Яндекс.Метрика